(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Metallo-porphyrin containing zirconium phosphonate thin films : structure and catalysis"

rv-fff.:-' 



METALLO-PORPHYRIN CONTAINING ZIRCONIUM PHOSPHONATE 
THIN FILMS: STRUCTURE AND CATALYSIS 



.• ^ \.v ■■*■>.. 



-. f 



■^ 



By 
CHRISTINE MARIE NIXON LEE 



A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL 

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT 

OF THE REQUIREMENTS FOR THE DEGREE OF 

DOCTOR OF PHILOSOPHY 

UNIVERSITY OF FLORIDA 

2000 



ACKNOWLEDGMENTS 

First, I would like to thank some of the teachers who have pointed me toward 
chemistry and supported me in this long educational adventure. I thank Mr. Roger 
Craig from Lexington High School, for being so excited about chemistry, and Dr. Jim 
McCargar, for being a tireless ambassador of general and physical chemistry, and for 
encouraging (pushing?) me to pursue research opportunities outside of Baldwin- 
Wallace College. I would also like to thank Dr. Gary Kosloski, who cared about my 
music even after I defected to the other half of the liberal arts and sciences. And to 
Dr. John West and Dr. William Samuels, I extend many thanks for accepting me into 
their research programs and giving me two unique and valuable perspectives on 
research outside of the academic envirorunent. 

I must thank Dr. Dan Talham for his patience, the limits of which, I have 
surely tested. I would also like to thank him for accepting a physical polymer 
chemistry convert and for teaching me to appreciate materials and surface chemistry. 
I would also like to thank him for his time and effort in encouraging all of us to be 
organized and effective speakers and writers. For that, I will be eternally grateful. 

I could not have completed this document or the research that it describes 
without the collaborative assistance from the Bujoli group in Nantes, France. 
Especially to Fabrice Odobel, who has contributed significant time and energy to 



u 



making and teaching others to make the porphyrins and ligands on which this 
dissertation is focused, Merci! 

Quick notes of thanks are also due to the Butler Polymer Research 
Laboratories for sharing their instrumentation and to Eric Lambers and the Major 
Analytical Instrumentation Center for allowing me to use the XPS. Also, thanks to 
Joe and Raymond in the chemistry department machine shop for working with me on 
designing and building the catalysis flow cells. 

Without many wonderful friends, my time at University of Florida would have 
been much less enjoyable. So, I thank Louarm Troutman, Tracey Hawkins, Dean and 
Annie Welsh, Denise Main, and Debby Tindall, and of course, Jen Batten and Leroy 
Kloeppner (and Alex) - the greatest of friends. I owe Jen and Leroy special thanks, 
not only for their efforts in editing this dissertation, but also for their support and love 
along the way. 

And lastly, my family. All the gratitude in the world goes to my mom and dad, 
Pat and Ron, and to my brothers Joe and John and my soon-to-be sister, JuUia. And to 
my grandmothers, Evelyn Nixon and Margaret Loeber, and in memory of my 
Grandpas Bill and Lee. I have been truly blessed. Thanks, too, to my new family, 
George and Agnes, Brian and Greg Lee who are the best second family I could 
imagine. 

To my husband, Larry, my best friend and biggest fan, I owe more thanks than 
can be expressed. 



m 



TABLE OF CONTENTS 

mss. 

ACKNOWLEDGMENTS ii 

LIST OF TABLES vii 

LIST OF FIGURES viii 

ABSTRACT xiii 

CHAPTERS 

1 INTRODUCTION 1 

1.1 Ultrathin Films 1 

1.1.1 Langmuir-Blodgett Films and Characterization 3 

1.1.2 Self-assembled Films 15 

1 .2 Hybrid Organic/Inorganic Ultrathin Films Based on Layered Solids 16 

1.2.1 Background 16 

1 .2.2 Self-assembled Films Incorporating Metal Phosphonate 
Binding 19 

1 .2.3 Metal Phosphonate Langmuir-Blodgett Films 20 

1 .2.4 Dual-Function Langmuir-Blodgett Films 23 

1.3 Background on Porphyrins 24 

1.3.1 Optical Behavior of Porphyrins 24 

1.3.2 Background on Manganese Porphyrins 33 

1.3.3 Immobilization of Porphyrins 35 

1.3.4 HeterocycUc Ligand Cocatalysts 36 

1.4 Dissertation Overview 38 

2 EXPERIMENTAL 42 

2.1 Langmuir-Blodgett and Self-Assembled Film Preparation Procedure 42 

2.1.1 General Langmuir-Blodgett and Self- Assembly Procedures 42 

2.1.1 Characterization 46 

2.2 Porphyrin Films 48 

2.2.1 Palladium Porphyrin Films 48 

2.2.2 Manganese Porphyrin Films 49 



IV 



2.2.3 Manganese Porphyrin/Imidazole Mixed Films 51 

2.3 Catalysis 53 

2.3.1 Catalysis using PhIO as an oxidant 53 

2.3.2 Catalysis using Peroxides as oxidants 57 



PALLADIUM PORPHYRIN CONTAINING ZIRCONIUM PHOSPHONATE 
THIN FILMS 58 

3.1 Background on Palladium Porphyrin Films 58 

3.2 Results 61 

3.2.1 UV-vis of Palladium Porphyrin Solutions 61 

3.2.2 Langmuir Monolayers of Palldium Porphyrins 63 

3.2.3 Langmuir-Blodgett Films 69 

2.3.3 Conclusions 79 



MANGANESE PORPHYRIN CONTAINING ZIRCONIUM 

PHOSPHONATE THIN FILMS 83 

4.1 Background 83 

4.2 UV-vis Behavior of MnTPPs 85 

4.2.1 Solution studies 85 

4.2.2 Langmuir Monolayers 93 

4.2.3 Langmuir-Blodgett Films of pure MnP4 95 

4.2.4 Self-assembled films of MnP4 100 

4.3 Conclusions 106 



INCORPORATION OF AN IMIDAZOLE LIGAND INTO MANGANESE 

PORPHYRIN CONTAINING ZIRCONIUM PHOSPHONATE 

THIN FILMS 109 

5.1 Background 109 

5.2 Solution Studies 113 

5.2.1 MnPO and MnP4 with ImH 113 

5.2.2 MnP4 and MnPO with ImODPA 115 

5.3 Film Studies 117 

5.3.1 Langmuir-B lodgett Films containing substituted MnP4 117 

5.3.2 Mn-porphyrins substituted into self-assembled films of 
ImODPA 124 

5.3.3 Self-assembling the MnP4 and ImODPA from a 

mixed solution 130 

5.3.4 Other methods for preparing ImODPA and MnP4 containing 
films 131 



5.3.5 Characterization of films containing MnP4 and ImODPA 

by XPS and ATR-IR 134 

5.4 Conclusions 141 

6 MANGANESE PORPHYRIN AND IMIDAZOLE CONTAINING 

ZIRCONIUM PHOSPHONATE THIN FILMS AS CATALYSTS 144 

6.1 Background 144 

6.2 Results 147 

6.2.1 Catalysis With PhIO as the Oxidant 147 

6.2.2 Catalysis Using H2O2 as the Oxidant 159 

6.3 Conclusions 163 

REFERENCES 164 

BIOGRAPHICAL SKETCH 171 



• 4 



wyv- V- 



VI 



LIST OF TABLES 
Table page 

3.1 UV-vis data from symmetric and alternating films of PdP4. Xmax. is given for 

monolayers, and interlayer thickness is given for multilayers of films 
transferred under a variety of transfer conditions 71 

3.2 UV-vis data from symmetric and alternating films of PdPl . X,max is given for 

monolayers, and interlayer thickness is given for multilayers of films 
transferred under a variety of transfer conditions 77 

6.1 Time dependence of epoxidation of cyclooctene using 40 |amol cyclooctene 

and 5 famol PhIO in ImL of solution. To the homogeneous reaction was 
added InmolofMnPO 151 

6.2 Conversion of cyclooctene to cycloctene oxide with 40 |amol cyclooctene 

and 5 i^mol PhIO in ImL of solution using MnP4 LB film 154 

6.3 Conversion of cyclooctene to cyclooctene oxide with varying cyclooctene 

to PhIO ratios in ImL of solution using MnP4 SA films over 24 hr 155 

6.4 Conversion of cyclooctene to cyclooctene oxide with 40 ^mol cyclooctene 
and 5 ^imol PhIO in ImL of solution and in films containing imidazole 157 



■V' e«. 



6.5 Comparison of blanks and homogeneous epoxidation yields in vials vs. in 

the reaction cells 158 

6.6 Conversion of cyclooctene to cyclooctene oxide with 400 fimol cyclooctene 

and 80 ^mol H2O2 in ImL of solution using imidazole and porphyrin 161 



vn 



LIST OF FIGURES 
Figure page 

1.1 Schematic of an isotherm and corresponding monolayer behavior 5 

1.2 Schematic of X-, Y-, and Z-type Langmuir-Blodgett multilayers 7 

1.3 X-ray diffraction diagram 9 

1.4 Illustration of ATR-IR Experiment 10 

1.5 Schematic of XPS experiment H 

1.6 Schematic of polarized UV-vis experimental beam directions 13 

1.7 Behavior of the oblique dichroic ratio versus an orientation parameter (P) 14 

1.8 Crystal structure of zirconium phosphate 17 

1 .9 Comparison between tradition LB films and metal-phosphonate LB films 20 

1.10 Schematic of formation of divalent or trivalent metal phosphonate films 22 

1.11 Structures of porphyrin-type molecules A) porphine B) free base 

porphyrin, and C) pthalocyanine 25 

1.12 UV-vis spectrum of a metallo-porphyrin(PdTPP) 25 

1.13 Outline of 1 6-member principle resonance structure of metallo-porphyrin 26 

1.14 Gouterman's four-orbital model 27 

1.15 Transition dipole moments in metallo-porphyrin 30 

1.16 Porphyrin chromophore interactions: The square represents the 

chromophore and its disecting axes. A) H-type or face-to-face 

aggregates; B) edge-to-edge aggregates; C) J-type or head-to-tail 

aggregates 32 

1.17 Suggested mechanism of olefin epoxidation catalyzed by MnTPP 35 



vui 



2.1 Schematic of Langmuir-Blodgett trough and monolayer 42 

2.2 Schematic of the three-step deposition process used for zirconium 

phosphonate films 45 

2.3 Schematic of catalysis cell, side view 54 

2.4 Schematic of catalysis cell, top view 55 

3.1 Structures of A) PdP4 and B) PdPl 59 

3.2 Schematic of Pd-porphyrin films formed: a) alternating ODPA/Zr/PdP, 

b) alternating ODPA/Zr/PdP:ODPA mixed film, c) symmetric 

PdP/Zr/PdP, d) symmetric PdP:ODPA/Zr/PdP:ODPA 60 

3.3 Solution UV-vis of Pd-porphyrins in CHCI3: A) PdP4, B) PdPl 62 

3.4 Solution UV-vis of PdP4 in EtOH and water compared to CHCI3 63 

3.5 Isotherms of PdP4, pure and mixed with ODPA (PdP4:0DPA), on a 

water subphase 64 

3.6 Reflectance UV-vis of PdP4 on water subphase 65 

3.7 Isotherms of PdPl, pure and mixed with ODPA (PdPl :ODPA), on a 

water subphase 66 

3.8 Reflectance UV-vis of PdPl on water subphase 67 

3.9 Reflectance UV-vis of 10% PdP4: 90% ODPA on a water subphase 68 

3.10 Mean molecular area vs. ratio of ODPA/Porphyrin: A) PdP4, B) PdPl 69 

3.11 Transmission UV-vis of PdP4 films transferred at high and low MMA. 

Absorbance scale corresonds to the film transferred at 300 A^ molecule'' 72 

3.12 UV-vis of SA PdP4 films rinsed in hot CHCI3 75 

3.13 Transmission UV-vis of films of PdPl transferred at high and low MMA 76 

3.14 Absorbance of Soret vs. time rinsed in hot CHCI3: A) PdP4, B) PdPl 78 

3.15 Illustration of orientation and packing of PdP 1 films transferred at high and 

low MMA 79 



IX 



3.16 Illustration of orientation and packing of PdP4 films transferred at high and 

lowMMA 80 

4.1 Structures of A) MnP4 and B) MnPO 84 

4.2 UV-vis of MnPO in CHCI3 87 

4.3 Solvent behavior of MnP4 in water, EtOH and CHCI3 88 

4.4 UV-vis concentration study of MnP4 in CHCI3: a) 10"^ M, b) 10"^ M 88 

4.5 MnPO in CHCI3 (1 x lO"' M) with ethylphosphonic acid: a) pure MnPO, 

b) 1 X 10"* M ethylphosphonic acid, c) 2 x 10"^ M ethylphosphonic acid, 

d) 3 X 10"^ M ethylphosphonic acid, e) pure MnP4 90 

4.6 Solution UV-vis investigation of MnP4's sensitivity to displacement of 

R-PO(OH)^ by chloride at 1x10"^ M. The arrows indicate the changes 

in the intensity of the peaks as the chloride concentration changes from 

0.0 M to 0.1 M while the concentration of MnP4 stay constant in CHCI3 93 

4.7 Isotherm of MnP4 on water subphase 94 

4.8 Reflectance UV-vis of MnP4 on water subphase 95 

4.9 UV-vis of MnP4 capping layers transferred onto ODPA/Zr at different 

surface pressures (indicated by the arrows) 96 

4.10 LB films of MnP4 transferred at A) 1 5 mN/m and B) 5 mN/m rinsed in 

CHCI3 98 

4. 1 1 MnP4 transferred by LB at 0.7 mN m'' and rinsed in CH3CN: A) 

transferred from a 0.5 mg mL'' solution 99 

4.12 MnP4 transferred from 0.1 M [Cf] aqueous subphase at 4 mN m"' 100 

4. 1 3 MnP4 self-assembled from EtOH/H20 and rinsed in CHCI3. The legend 

indicates the spectra after rinsing, after being left overnight and the 

rinsed again over a three day period 101 

4.14 SA MnP4 films with rinsing in hot CH3CN 102 

4.15 UV-vis response of a SA MnP4 film during rinsing with hot CH3CN 103 

4.16 UV-vis of MnP4 self-assembled films before and after rinsing in hot EtOH 104 

4.17 MnP4 self-assembled from a 0.1 M chloride solution 105 



4. 1 8 XPS of MnP4 SA film. The insert is an enlarged view of the same spectrum 

between 200 and 80 eV 106 

5.1 Structures of A) MnP4, B) MnPO, C) ImODPA and D) ImH 110 

5.2 Simplified Schematic of MnP4 and ImODPA incorporation in films 11 1 

5.3 Solvent response of A) MnPO and B) MnP4 to ImH 114 

5.4 Solvent response of A) MnPO and B) MnP4 to ImODPA. Legends indicate 

the molar ratio of MnP to ImODPA 116 

5.5 UV-vis of ODPA/Zr/HDPA, SA MnP4 film rinsed in hot CHCI3 119 

5.6 MnP4 substituted onto ImODPA:HDPA LB films after CHCI3 rinsing 121 

5.7 MnP4 substituted onto a 25% ImODPA/HDPA film, rinsed in room 

temperature and hot CHCI3 122 

5.8 UV-vis of an ImODPA/ MnP4 film after drying 123 

5 .9 MnPO attached to a 25% ImODPA/HDPA LB film and rinsed in hot CHCl 1 24 

5.10 MnP4 substituted onto a pure ImODPA SA film 126 

5.11 Reversibility ofthechloride/phosphonic acid binding 127 

5.12 MnP4 substituted film rinsed in chloride and t-butylamine solutions 128 

5.13 MnP4 substituted from a 0. 1 M CI- solution onto an ImODPA layer, and 

compared to an MnPO solution with ImH binding 129 

5.14 ImODPA/MnP4 self-assembled from 70/30 mixture and rinsed in hot 

CHCI3 131 

5.15 ImODPA substituted into a MnP4 LB film transferred at 10 mN m"' 132 

5.16 LB film of MnP4/ImODPA transferred from a 25/75 mixture on an aqueous 

subphase, pH 11.3 I33 

5.17 XPS muhiplex scan over the Nis region of A) ImODPA, and B) MnP4 self- 

assembled films. The dashed line represents the Gaussian peak fit 134 



XI 



5.18 XPS multiplex scan of Nl s region of ImODPA/MnP4 film self-assembled 

out of 70/30 CH2CI2 solution. The dashed lines represent the Gaussian 

peak fits 136 

5.19 XPS multiplex scan of ImODPA/MnP4 film self-assembled from a 70/30 

mixture in EtOH/HiO 136 

5.20 ATR-IR of ImODPA SA film 138 

5.21 Increase in absorbance intensity of 29 1 8 cm"' peak in ImODPA with 

SAtime 139 

5.22 ATR-IR of alkyl region of: A) MnP4 substituted on a 1 00% ImODPA base 

capping layer, B) MnP4 substituted on a 25% ImODPA base 

capping layer 141 

6.1 SA MnP4 film before and after 24 hr. catalysis run with 40:5:20 

cyclooctene: PhIO:decane in CH2CI2 148 

6.2 SA MnP4 film before and after 2 hr. catalysis run with 40:5:20 

cyclooctene: PhIO:decane in CH2CI2 149 

6.3 UV-vis of MnP4 film SA from chloride containing solution used in 

catalysis with PhlO after 6 hr 150 

6.4 Bleaching of MnPO in homogeneous catalysis reaction with PhIO 152 

6.5 MnP4 LB film before and after 24 hr catalysis reaction 153 

6.6 SA ImODPA/SA MnP4 studied with PhIO for epoxidation of cyclooctene 156 

6.7 SA ImODPA/SA MnP4 studied in the epoxidation of cyclooctene 

using H2O2 159 

6.8 SA ImODPA/SA MnP4 after rinsing and after 24 hr in catalysis reaction 

with excess H2O2 161 

6.9 SA ImODPA/SA MnP4 film with catalysis using 8 ^mol cyclooctene 

to 0.2 ^imol H2O2 162 



Xll 



Abstract of Dissertation Presented to the Graduate School 

of the University of Florida in Partial Fulfillment of the 

Requirements for the Degree of Doctor of Philosophy 

METALLO-PORPHYRIN CONTAINING ZIRCONIUM PHOSPHONATE 
THIN FILMS: STRUCTURE AND CATALYSIS 

By 

Christine Marie Nixon Lee 

August 2000 

Chairperson: Daniel R. Talham 
Major Department: Chemistry 

Thin films containing mono- and tetra-phosphonic acid palladium tetraphenyl 
porphyrins and tetra-phosphonic acid manganese tetraphenyl porphyrins, PdPl, PdP4, 
and MnP4, were prepared by both Langmuir-Blodgett (LB) and self-assembly (SA) 
techniques. Within the hydrophilic regions of these films was incorporated a 
zirconium phosphonate network which lent significant stability and flexibility of 
preparation to these films. 

In the LB films of the palladium porphyrins, it was found that the mono- 
phosphonic acid porphyrins aggregated under all film preparation conditions and in 
solutions at high concentrations. However, the chromophore aggregation could be 
controlled in the tetra-phosphonic acid porphyrins when the films were transferred at 
mean molecular areas greater or near the mean molecular area of the chromophore 
itself Self-assembling the PdP4 was another means of controlling the chromophore 



xni 



interaction in the films. Because chromophore aggregation was expected to inhibit 
catalysis, film preparation conditions were sought in order to avoid this. 

LB and SA films of pure manganese porphyrins were successfiilly prepared by 
a number of different methods. Aggregation appeared insignificant when transferred 
at high mean molecular areas and when the films were self-assembled. The pure 
manganese porphyrin films were successfiil at catalyzing the epoxidation of 
cyclooctene using iodosylbenzene as the oxidant. 

To activate the porphyrin for catalysis with peroxide oxidants, a heterocyclic 
ligand was also incorporated into the manganese porphyrin containing films. The 
heterocyclic ligand used was an imidazole substituted with an octadecylphosphonic 
acid chain. Both the manganese porphyrin and the imidazole amphiphile were 
tethered to a zirconium phosphonate network, first for ease of film synthesis and 
second, to stabilize the film for use in the catalysis reactions. 

Though significant catalyst degradation has been reported in homogeneous and 
alternative heterogeneous catalysis studies, the manganese porphyrin and imidazole 
containing zirconium phosphonate films were generally more resistant to degradation 
under catalysis conditions. The stability of the films toward epoxidation conditions 
has led to easily recyclable catalysts. 



XIV 



CHAPTER 1 
INTRODUCTION 



1.1 Ultrathin Films 



The study of ultrathin films, especially monomolecular thick films, enables the 
study of two-dimensional systems and allows the simplification of complicated 
thermodynamic behaviors. Recent interest in monolayers and multilayers focuses on 
the many potential applications of organized and functional thin films, which include 
optoelectronics, '-3 coatings,'*-'^ chemical sensors,' '6.8-10 and heterogeneous 
catalysts.' i-'3 In order to prepare these organized and essentially two-dimensional 
structures, the Langmuir-Blodgett (LB) and self-assembly (SA) techniques have been 
developed. .' 

The study of monolayer thick films began long before the early twentieth 
century investigations of Irving Langmuir and Katharine Blodgett. It is believed that 
centuries ago drops of oil were used to calm waves in ponds and other small bodies of 
water.14,15 Benjamin Franklin studied monolayers of oil on the surface of a pond. 
Agnes Pockels pioneered the study of a monolayer on the water surface in the 
laboratory environment, and is credited with building the first trough. '5 However, the 
first systematic study of monolayers of amphiphilic molecules on aqueous surfaces 
began with Irving Langmuir's studies at GE Laboratories and, hence his name is 
associated with a fundamental method of preparing organized, monomolecular thick 



1 



IT J •' 



films. 16 His associate, Katherine Blodgett first transferred these films fi-om the water 
surface onto solid supports. •'^ 

The SA technique was first described in scientific reports in the late 1940's and 
mid 1950's.>8,i9 x^jg technique relies on surface-acfive molecules and the appropriate 
surface being placed in contact with one another through a solvent medium. The films 
prepared by this method tend to be more stable than traditional LB films because of 
the types of surface interactions that drive their formation. 

The traditional LB technique employs different surface interactions depending 
on the method of transfer. First, as in the case of hydrophilic-hydrophilic transfers of 
neutral amphiphiles from a pure aqueous subphase, the interactions are primarily 
hydrogen bonding in nature. Ionic bonding is commonly observed in the case of 
hydrophilic-hydrophilic transfers from a metal ion-containing subphase. In the case of 
hydrophobic-hydrophobic transfers, van der Waals interactions are involved. During 
LB depositions, the film is usually physisorbed to the surface implying some through- 
space interaction, while in SA films, the molecules are adsorbed to the surface through 
a chemical bond. The SA method relies on the formation of covalent bonds between 
the surface-active component of the molecule and the solid substrate often resulting in 
more stable films.^o 

In LB films, the hydrophilic head group typically dictates the area the molecule 
fills in the interfacial region. The alkyl groups, therefore, adjust to maximize the van 
der Waals contacts leading to organized packing within this region.^' Unfortunately, 
control over the film packing and organization achieved by the LB technique is absent 
in the SA method. The molecular organization in SAMs (self-assembled monolayers) 
is dictated solely by the geometry of the active sites on the surface. However, in LB 
films, the pressure and area of deposition can be selected to deposit a particular phase 
of the monolayer and to influence the transferred film organization. An understanding 



of the mechanics of both the LB and SA processes is important in appreciating thin 
film research and its appUcation to many areas of science.20 

1.1.1 Langmuir-Blodgett Films and Characterization 

1.1.1.1 Langmuir monolayer formation: the isotherm. The Langmuir 
monolayer is achieved by placing droplets of an amphiphile solution, in a volatile 
solvent such as CHCI3, on the aqueous subphase in such a way as to uniformly spread 
the compound on the surface. Typically, the presence of an amphiphile works to 
decrease the surface tension, and the difference is defined as the surface pressure. In 
Equation 1.1, n is the two-dimensional surface pressure typically measured in mN m', 
Yo is the initial surface tension of the subphase and Yf is the surface tension of the 
subphase and film.22.23 

n = r„-r/ (1.1) 

The record of the monolayer formation on the water surface, the n vs. area (A) 
isotherm, depends on the change in the surface pressure with a change in the mean 
area per molecule (MMA) on the surface. '^'22 jhe surface pressure is measured using 
a sensitive microbalance, such as a Wilhelmy balance, while the area is computer 
controlled using movable barriers, which define the monolayer boundaries. 

Upon room temperature spreading, the monolayer is at a very low density and 
behaves like a two-dimensional gaseous phase (Figure 1.1). In this phase, the 
molecules are theoretically not in constant contact with one another, although 
aggregation may occur depending on the affinity of the amphiphilic molecules for one 
another. There are random collisions, but there is no appreciable increase in surface 



pressure because, in effect, there is not a true monolayer, and the presence of the 
amphiphile has virtually no effect on the surface tension of the subphase.20 There is 
some debate on the existence of a two-dimensional gaseous state; some researchers 
claim that there is always aggregates formed, which interact within the gaseous 
phase.23 

As the monolayer is compressed, the molecules will come in contact with one 
another, and an observable rise in n occurs. In the initial region of noticeable pressure 
increase, the molecules are colliding, but the film is in a fluid-like state. The 
molecules have no long-range orientational order and they are not close-packed or 
organized. This region is often called the liquid-expanded state (LE) (Figure 1.1). 
Within the LE phase, the alkyl chains have many degrees of freedom, and gauche 
conformations are observed within the chains. 

The further compression of long chain fatty acid amphiphiles leads to a more 
crystalline and organized monolayer. In this phase, the hydrophobic tails of the 
amphiphiles adopt an overall orientational order that is maintained within the film 
domains. This orientational order seeks to balance the van der Waals interactions 
between the alkyl chains with the pressure applied by the barriers. Within this region, 
referred to as the liquid-condensed phase (LC) (Figure 1.1), the slope of the isotherm 
is very steep, meaning that the pressure goes up rapidly without much change in the 
MMA. In some cases, the structure of the amphiphiles may prohibit the formation of a 
close-packed monolayer regardless of the pressure, and therefore, the monolayer enters 
and remains in the LE phase. ' ^ '• 

When the monolayer cannot compress any further, additional applied pressure 
from the barriers causes the monolayer to fold either over or under itself forming 
collapsed regions. The collapse point is identified as the first point of deviation from 
the linear slope of the LC region of the isotherm. '5.22 



1=1 



C. Liquid Condensed State 




^.. f^ 



u. 



• ^ 



A. Gaseous State 



MMA (A^ molecule"^) 



Figure 1.1: Schematic of an isotherm and corresponding monolayer behavior. 



1.1.1.2 Langmuir monolayer characterization: creep and hysterisis tests and 
reflectance spectroscopy. Although the isotherm is very important in identifying the 
general behavior of a monolayer film, it does not give specific information about such 
things as the monolayer stability, alkyl chain orientation, or aggregation of the 
amphiphiles. These more specific ideas of monolayer behavior can be discerned from 
experiments such as creep and hysterisis tests, reflectance UV-vis, fluorescence 
microscopy, Brewster angle microscopy, and surface potential measurements,^^ to 



name a few. 15.22 x^e methods applying to the films discussed in this dissertation will 
be described here. 

Creep tests provide information on monolayer quality and can be recorded in 
two ways. First, the monolayer is compressed to a certain pressure and then that 
pressure is maintained by the barriers expanding or compressing as necessary while 
recording the change in area with time. Second, the monolayer is compressed to a 
defined area, which is maintained by stationary barriers, and the pressure is monitored 
over time. Stable monolayers will show a static surface pressure or little movement in 
the barriers after the desired pressure is reached. Unstable monolayers go through 
constant and sometimes drastic rearrangements, which force contraction or expansion 
of the barriers. For example, the hydrophobic nature of the amphiphile may be 
insufficient and the amphiphile may dissolve into the subphase forcing the barriers to 
compress to maintain FI. Also, the vapor pressure of the amphiphile may lead to their 
evaporation, causing the surface pressure at constant area to decrease or the barriers to 
move forward to hold constant pressure in proportion to the instability of the film. 
Alternatively, the amphiphiles may have a strong affinity for one another, leading to 
agglomeration and either causing an anomalous change in the surface pressure at 
constant area or forcing the barriers to work to maintain the surface pressure.22 

Hysterisis studies monitor the effect of the monolayer stability on the 
reproducibility of the isotherm upon compression and decompression. If the 
amphiphiles tend to aggregate, the isotherm will not retrace its compression curve in 
its decompression cycle. From creep tests and hysterisis experiments, the ability of 
the monolayer to hold its form and to possibly be transferred can be ascertained.24 

Reflectance UV-vis spectroscopy is used to understand the optical behavior of 
monolayers of chromophore-containing amphiphiles on the water surface. One 
method for studying the Langmuir monolayer by reflectance UV-vis involves placing 



a mirror on the base of the trough and reflecting a beam through the monolayer onto 
this mirror, and back into a detector. These studies can help determine the onset of 
aggregation in films with such a tendency.25.26 

>.1.1.3 Langmuir-Blodgett film formation. LB films are formed by vertically 
transferring Langmuir monolayers from the water surface onto a solid support. There 
are three common vertical dipping techniques, which form X, Y, and Z-type films 
(Figure 1 .2). 15-23 In X-type LB films, a Langmuir monolayer is transferred onto a 
hydrophobic substrate in such a way as to maintain head to tail type interactions. In Z- 
type LB films, the monolayer is transferred onto a hydrophilic substrate also forming 
head to tail interactions. X- and Z-type films can be prepared on a specially designed 
trough, which allows one stroke to be made through a monolayer and the next to be 
made through a clean subphase. However, some amphiphiles have a preference for 
this type of interaction and upon regular dipping, these structures form spontaneously. 
Y-type multilayers are most common and can be prepared on either hydrophilic or 
hydrophobic substrates. Y-type multilayers are typically the most stable due to the 
strength of the head-head, tail-tail interactions. '^ 



• • • 

• • • 

• • • 

• • • 

• • • 

• • • 

• • • 

• • • 

• • • 

• • • 

• • • 

• • • 



X-type 



Z-type 



Y-type 



Figure 1.2: Schematic of X-, Y-, and Z-type Langmuir-Blodgett multilayers. 



1 ■ 1 ■ 1 .4 Langmuir-Blodgett film characterization. The quality of the transferred 
film is first indicated by the transfer ratio, which is a measure of the change in the area 
of the monolayer versus the area of the substrate coated by the monolayer. A transfer 
ratio of unity indicates that the monolayer is transferred with the same area per 
molecule that it had on the water surface. This "perfect" transfer ratio assumes that the 
monolayer on the water surface was stable and was not reorganizing significantly 
during transfer. A consistent deviation from unity could imply a change in 
organization upon transfer; however, if the transfer ratio is irregular, the transferred 
film is probably poor quality. ^ ^ 

There are many analytical techniques used to study transferred films. Film 
characteristics typically of interest are thickness, interlayer spacing, molecular 
orientation and packing, film coverage, surface topology, chemical composition, and 
optical and magnetic properties. The techniques used to study these parameters are 
well described in the literature.^o 

X-ray diffraction is a reliable technique to probe interlayer spacing, and from 
interlayer spacing, film thickness can be inferred. The X-rays are essentially reflected 
fi-om planes of higher electron-density. The interaction of X-rays with the crystalline 
planes can be described by Bragg's law (Equation 1 .2): 

.^^■■:-\ nA = 2dsme (1.2) 

where n is an integer. A, is the wavelength of the radiation, d is the interlayer spacing, 
and is the angle of incidence and reflection of the beam.27 

Ideally, there should be a significant difference between the electron density of 
the head group and that of the hydrophobic region allowing X-ray diffraction peaks to 



be observed in LB films. The d-spacing, which quantifies the periodicity between 
planes of high electron density, therefore, is the measure of the distance between head 
groups. This technique is very sensitive to long-range periodicity, and many narrow 
00/ peaks are typically indicative of a well-defined layered architecture (Figure 1 .3).23 




d-spacing 



Figure 1.3 X-ray diffi-action diagram. 



To study the chemical make-up of the surface, attenuated total reflectance 
FTIR (ATR) and X-ray photoelectron spectroscopy (XPS) are employed. ATR studies 
involve transferring a film onto a crystal, such as parallelogram-shaped germanium or 
silicon crystal with ends cut at 45° angles. This is an ideal method for recording the IR 
spectra of films because it allows the film to be sampled many times due to the 
internal reflection of the IR beam through the crystal. ATR also provides information 
about the surface coverage and packing modes. Further, polarized studies using this 
technique can give information about the film organization and orientation (Figure 
1.4). '5 Background information pertaining to films studied by ATR-IR, allows 



10 



elucidation of information about films containing new amphiphiles. For example, by 
comparing the areas of the alkyl stretches of a new amphiphile to that of well- 
understood fatty-acid films, an idea of the transfer quality can be obtained. This tool 
is particularly helpful in studying monolayers with transfer ratios varying from unity. 

ATR-FTIR can also indicate the packing nature of the alkyl chains. If the 
chains are in a mostly trans configuration and close-packed, the asymmetric CH2 
stretch (v^CHj) will occur at 2918 cm'' and have a full width at its half maximum 
(FWHM) of ca. 20 cm-'. The symmetric CH2 stretch (v.CHj) will occur at ca. 2852 
cm-'. If there are a significant number of gauche interactions in the chains, the V3CH2 
and V5CH2 stretches will shift to higher energies (ca. 2924 cm-' for the v^CHj and ca. 
2856 cm-i for the v.CH^ stretches).28.29 



6666666666666666 




QQQQOQQQOOOQO 



IR Beam 



Figure 1.4: Schematic of the ATR-IR Experiment. 



XPS is a method used to determine the elemental make-up, and possibly, 
atomic proportions within a film based on the photoelectron effect. XPS measures the 
energy of an expelled electron as the surface is bombarded with a monochromatic X- 
radiation source (Figure 1.5). When a high-energy source is applied, the kinetic energy 



11 



of the emitted electrons can be related to the excess energy and to the strength of the 
electron's binding by Equation 1.3, which is the Einstein photoelectric law:27 

E,„=hv-e<p-E, (1.3) 

where Et is the binding energy, e is the charge of the electron, h is Planck's constant, v 
is the frequency of the radiation, and (f) is the work function corresponding to the 
minimum energy required for ejection of an electron. 



Monochromatic X-ray 
Beam 



\ 



Detector 
Ekin =hv-Eb 




Figure 1.5: Schematic of XPS experiment. 

The XPS spectrum plots electron counts versus binding energy. Binding 
energies are unique to each element present in the film, as well as to that element's 
chemical enviromnent and oxidation state. Therefore, from a scan over a wide range 
of binding energies, called a survey scan, XPS results can be used to define the 
elements present in the sample. A more extensive scan, or a multiplex scan, over a 
narrow range of binding energies can clarify peak splitting, which can be assigned to a 
change in the chemical environment or oxidation state. Offord, et. al., in their study of 
a Ru-porphyrin linked to a thiol on gold SA film containing a percentage of imidazole 
terminated thiol surfactants, observed two peaks within the N,, region of the XPS 



12 



spectrum. These two peaks were associated with the different nitrogen environments 
in the imidazole and porphyrin, clearly indicating that both species were present in the 
mixed film. 30 

The intensities of the XPS peaks can be used to determine the relative ratios of 
the elements present, and can indicate the type of crystalline lattice formed. However, 
the intensities of the XPS peaks are sensitive to many parameters such as the element's 
electron escape depth, which can complicate the determination of the elemental ratios. 
In a given sample, the observed relative peak intensities are compared to a calculated 
value based on Equation 1.4: 



^aT^^^P 



L = 



-d. 



A,,^(sin6') 



^"E^^P 



-d„ 



\^(sin^) 



+ 



/;2]exp 



-d.. 



^...flCsin^) 



+ . 



(1.4) 



where I^ is the relative intensity of element A, 1^°° is the atomic sensitivity factor, d„ is 
the overlayer thickness, is the incident angle of the X-ray beam, and X^ is the 
inelastic mean free path. The inelastic mean free path represents the distance over 
which 60% of the electrons can travel before inter-electron collisions lead to a loss of 
energy3i and is defined by :32 



A,=10[49/(£^,/) + 0.11(£,„,r](A) 



(1.5) 



UV-vis spectroscopy reveals a film's optical behavior. Typically, films are 
transferred onto glass or quartz substrates and transmittance studies are performed. 



13 



Using polarizers and stages that allow careful placement of the substrates at different 
angles of incidence to the beam, the orientation of the chromophores within the films 
can be determined. By comparing the absorbance intensity in the s- and p- 
polarization, a dichroic ratio can be calculated using Equation 1 .6: 



D = 



(1.6) 



Orientational order within the plane of the film and substrate is obtained by looking at 
results from 0° and 90° (s and p) polarization as the beam is normal to the surface (0° 
angle of incidence). Studying the polarization at higher angles of incidence (typically 
45°) enables the determination of the orientation out of the plane (Figure 1 .6). 

* V- 

\ \V' ' Qo polarization 

90° polarization 




450 orientation 



0° orientation 



Figure 1.6: Schematic of polarized UV-vis experimental beam directions. 



14 



The dichroic ratio can be used to determine the chromophore orientation in a 
film. In the case of films containing porphyrin chromophores, Orrit et al. determined 
that the dichroic ratio can be translated into an orientation parameter (P) using the 
graph shown in Figure 1.7,25 and hence, an orientation angle (6) can be established 
using Equation 1.7: 



P = (cos^0) 



(1.7) 



represents the angel between the surface normal and the chromophore's molecular 
plane. As is often observed within the plane of a film containing porphyrins, if D is 
unity, there is no preferred orientation. If D = 1 .5, P = and therefore, 9 = 90 ° as 
measured from the surface normal. 




0.2 0.4 0.6 0.8 

Orientation Parameter (P) 



Figure 1.7: Behavior of the oblique dichroic ratio versus an orientation parameter 
(P).25 



15 



Porphyrin containing films often have oblique dichroic ratios of approximately 
1.5, corresponding to the chromophores lying parallel to the substrate. For example, 
Zhang et al., obtained such a result in LB films containing a free base tetraphenyl 
porphyrin either pure or mixed with stearic acid, where the porphyrin was the 
hydrophilic head group with long alkyl substituents.33 

1.1.2. Self-Assembled Films 

Zisman introduced SAMs to the literature little more than 50 years ago. His 
studies involved self-assembling long-chained alcohols onto a glass surface using 
hexadecane as the inert solvent. 18 This study showed that films prepared in this 
manner had wetting properties similar to those seen in films prepared by the LB 
technique. 

Sagiv et al. studied octadecyltrichlorosilane (OTS) on hydroxylated surfaces, 
such as glass, to form a siloxane polymer. Multilayers were produced if the 
amphiphile was terminated at both a and co positions by surface-active groups. 15,34 
When a and co positions were two different ftinctionalities, subsequent dipping in self- 
assembly solutions produced non-centrosymmetric films. Unfortunately, studies have 
shown that small defects in the early layers can magnify upon multilayer formation 
such that nearly all order breaks down by the tenth layer.20 The SA of OTS is now 
commonly used for hydrophobisizing glass slides for LB substrates. 

The chemisorption of thiols on gold was initiated by Nuzzo and AUara^s and 
continued by Porter,36 Whitesides,37 and others. Alkyl thiols, in which the chain 
lengths range from one carbon to over twenty, have been studied. Closely packed 
layers were observed when the chain length exceeded eleven carbons.36 The gold 
surface used in these studies was formed by vacuum evaporation onto cleaved alkali- 



16 



metal halide surfaces.37 Many of the surfactants studied by self-assembly did not form 
stable monolayers on aqueous subphases, and therefore, could not be studied by the 
LB method. However, these surfactants were easily studied by the SA method. 

1.2. Hybrid Organic/Inorganic Ultrathin Films Based on Layered Solids 

1.2.1. Background 

A new class of LB films has been developed, which incorporate an inorganic 
metal phosphonate network into the polar region.28.38-4i These films, which can 
contain a variety of organic groups and metals, were inspired by and show analogous 
behavior to their solid-state metal phosphonate analogues. In addition, these films are 
more stable than typical fatty-acid based LB films, and the inorganic lattice provides 
potential function.'*^ 

The metal phosphonate solid-state materials are attractive because they can be 
prepared at low temperatures from aqueous solutions. Further, the structures can 
provide a model system to which the film properties can be compared. Due to the 
structure of the metal lattices, which directs the film formation, the orientation and 
packing of the alkyl region is predictable. 

Metal phosphonate materials are especially interesting due to their potential 
applications in the areas of sorbents43 and catalysts,44 and because of their layered 
structures, these materials can be used as intercalation compounds.45-49 x^g interest in 
the metal phosphonates was sparked by their potential as inorganic ion exchange 
materials;50-52 however, the organic region can also be modified and functionalized, 
providing a straightforward method for preparing a wide variety of materials. 



17 



1 .2. 1 . 1 ■ Zirconium Phosphonate solids. Clearfield published early work on 
metal phosphonates and phosphates in the 1960's.53 The focus at this time was on the 
zirconium solids which form two preferred phases, a and y, which have the 
compositions Zr(HP04)2-H20 and Zr(P04)(H2P04)-2H20, respectively. In these 
layered materials, a two-dimensional metal lattice is formed and separated from an 
adjacent metal lattice by the organic layer in the phosphonates or by the hydrogen 
bonds from the fourth hydroxy site in the phosphates. The interlayer area in these 
solids forms a possible domain for intercalation of inorganic materials such as 



ammes. 



53 



.OH 




Figure 1.8: Crystal structure of zirconium phosphate.53 



The crystal structure of the Zr-phenylphosphonate solid was determined in the 
1 980's and found to form a structure similar to the a-phase. Subsequent studies 
revealed that any alkyl or aryl group whose area was under 24 A2 would form an 
identical metal lattice structure while only the interlayer distance changed. In the n- 



18 



alkyl phosphonates, it was found that there is a tilt angle in the chains of between 55° 
and 60°. This tilt allows for maximization of the van der Waals forces between the 
chains even within the solid-state materials. Though the structures formed are dictated 
by the geometry of the inorganic lattice, the hydrophobic region does rearrange to 
balance these strong forces with the maximization of their overlap.53 A change in the 
organic group can lead to very different structures and properties of the solids. Bulkier 
groups may form new crystal structures or have three-dimensional metal lattice 
formation. Also, the different organic groups can impart different function to the 
films. 

1 .2. 1 .2. Divalent and trivalent metal phosphonate solids. After extensive 
research on the zirconium phosphonate solids, interest branched to divalent and 
trivalent metal phosphonates. The poor solubility of the zirconium phosphonate solids 
in most standard solvents meant achieving single crystals was difficult; therefore, most 
of the crystal structure data was achieved from powder X-ray diffi-action patterns. 
However, di- and trivalent metals tend to be soluble in acidic solutions, allowing 
single crystals to be obtained by slowly changing the solvent pH or the metal ion 
concentration. 

Metal phosphonate materials have been prepared with a variety of different 
divalent metals such as Mg^\ Mn^^ Zv^\ Q,^\ and Cd^^54-56 prom the crystal 
structures, it was determined that the composition of the divalent series metal 
phosphonates is M"(03PR)-H20 for Mg, Mn, Zn, Ca and Cd. In these materials, 
layers of the metal atoms are octahedrally coordinated by five phosphonate oxygens 
and one water molecule, with each phosphonate group coordinating four metal atoms 
making a cross-linked M-0 network. A second structure, the orthorhombic 
M"(H03PR)2, was observed for the Ca phosphonates. A structural exception to the 



19 



above divalent series is Cu(II).''8.57 The Cu atoms in these phosphonate materials are 
five coordinate and form a distorted tetragonal pyramidal geometry. 

Mallouk has prepared a series of lanthanide phosphonates. The structure of 
these materials is given as Ln(III)H(03PR)2, where Ln represents La, Sm, or Ce. The 
lanthanide-series phosphonates are more soluble than the zirconium solids, but less 
soluble than the divalent materials. Therefore, single crystal data was not easily 
obtained.58.59 

ATR-IR provides a facile method for characterization of the metal phosphonate 
lattice formation. Vibrational modes assigned to the phosphonate are extremely 
sensitive to the mode of metal binding. Thomas et al. have assigned the VJCH2) 
vibrational frequencies for the divalent metal-phosphonates to be in the range of ca. 
1050 - 1 100 cm-' where the V3(CH2) stretches occur ca. 970 - 990 cm-'.60 Each metal 
phosphonate material has characteristic stretches in these regions. 

1.2.2. Sel f-Assembled Films Incorporating Metal Phosphonate Binding 

After Sagiv's work with self-assembling films of octadecyltrichlorosilane 
(OTS), it seemed a natural step to translate the formation of the thermodynamically 
stable and insoluble layered metal phosphonate solids into ultrathin films. Self- 
assembly was the first technique employed to produce metal phosphonate thin films. 
Mallouk and coworkers formed the metal phosphonate self-assembled films by first, 
exposing a silicon or gold surface to an appropriate template forming alkyl-mercaptan 
that was substituted with a terminal phosphonic acid.6i.62 This phosphonic acid was 
active toward the metal salt solution in which the substrate was then dipped. After 
metallating the surface, the substrate was dipped in a solution of an a, co- 
bisphosphonic acid, which left another phosphonic acid on the surface to be 



20 



subsequently metallated, and the cycle continued until multilayered films were 
fabricated. 

The self-assembly of metal phosphonate films is made possible by a very 
strong attraction between certain metal ions, particularly the tetravalent metals such as 
Zx'^, for the phosphonate groups of alkyl phosphonic acids.6i-63 However, the 
individual metal salts and the phosphonate are themselves soluble. This particular 
affinity between metal and phosphonate, makes possible the formation of 
monomolecular layers during each step of the cycle and the ability to assemble 
controlled multilayers. 

1.2.3. Metal Phosphonate Lanemuir-Blodgett Films 

Two methods of film formation have been employed to incorporate the metal 
phosphonate lattice into the polar region of LB films, one for the divalent and trivalent 
metals which are soluble in acidic media, and one for the tetravalent and some 
trivalent metals which are insoluble even at low pH. A schematic comparing 
traditional LB films to metal-phosphonate LB films is shown in Figure 1 .9. 

Hydrophobic 

A ^^A^^ region ^ i^i^^^^ 

' ' I ' I phosphonate I I | | | 

region 

A B 

Figure 1.9: Comparison between A) traditional LB films and B) metal-phosphonate 
LB films. 



21 



The zirconium metals have such a high oxophilicity for the phosphonate 
oxygens, that when a phosphonic acid amphiphile is spread on the surface of a 
zirconium cation containing aqueous subphase, the monolayer crystallizes before it 
can be transferred. Therefore, a three-step deposition procedure has been developed. 
A phosphonic acid containing monolayer is formed on the surface of a pure water 
subphase and transferred onto a hydrophobic substrate. Onto this phosphonic acid 
surface, a layer of zirconium is assembled, followed by the transfer of a second LB 
monolayer containing a phosphonic acid. This three-step deposition technique will be 
described in detail in Chapter 2.28.38 

Advantages of this three-step technique include the fact that at the hydrophilic 
stage, the monolayer is stable and can be independently characterized by ATR-FTIR 
or XPS, etc. Second, this method allows the formation of alternating films in which 
the template and capping layers do not have to be formed of the same amphiphile. The 
option of forming alternating films is important because some amphiphiles do not 
transfer on the down stroke but will transfer onto an ODPA template layer. A 
disadvantage of the three-step deposition of the phosphonate films occurs at the self- 
assembly of the zirconium lattice. The self-assembly of the zirconium onto the ODPA 
template causes the metal phosphonate lattice to be amorphous, whereas in a one-step 
deposition, the metal lattice is crystalline. 

An alternative technique of film formation is employed for the divalent and 
trivalent metals.''! In these cases, the metal salts are dissolved in the aqueous 
subphase, the phosphonic acid monolayer is formed on the surface, and the 
hydrophobic substrate is dipped dovm and then up through the same compressed 
monolayer forming a complete metal-phosphonate layer or an LB bilayer. The 
crystallization of the metal-phosphonate lattice occurs on the slow upstroke of the film 
through the monolayer (Figure 1.10). 



22 



In the one-step method, the pH of the subphase is as crucial to successful 
lattice formation as it is in the formation of the solids in aqueous solutions. If the pH 
is too high, the affinity of the metals for the deprotonated phosphonates will be too 
high and crystallization of the lattice will occur in the Langmuir monolayer rather than 
upon transfer. As with the zirconium films, these films will be too rigid to be 
successfially transferred. 



liiiii^v^imi 



• ^ • »■ 

.• « • . 



M=» 



e • ^ 



t 



liiiiil^.: 



t 







Figure 1.10: Schematic of formation of divalent or trivalent metal phosphonate films. 



If the pH is too low, the phosphonate will remain completely protonated, and the films 
will transfer without metal binding. Fortunately, the pH effects on the crystallization 
of the monolayer have signatures in the isotherm behavior.64 When the pH is too high, 
the rigid monolayer gives an erroneous but characteristic isotherm that has a much 



23 



higher onset and shallower incline. We believe this isotherm behavior is due to the 
rigid films causing deflections in the Wilhelmy balance rather than showing an 
increase in surface pressure. 

1 .2.4. Dual-Function Langmuir-Blodgett Films 

After the extensive characterization of simple alkyl metal phosphonate LB 
films, there was interest in incorporating fiinction into the organic region that might be 
paired with properties in the inorganic lattice to form a "dual fimction" LB film. As 
models, phenoxy and biphenoxy alkyl phosphonic acids were prepared, and divalent, 
trivalent, and tetravalent metal phosphonate films were studied. 65,66 Additionally, 
films containing azobenzene-derivatized phosphonic acid amphiphiles were 
synthesized, and metal phosphonate films were formed also with divalent, trivalent, 
and tetravalent metals.67 These results prove that larger organic groups can be 
incorporated into the metal phosphonate LB films while maintaining the integrity of 
the inorganic lattice structure. 

Potential applications of dual functional metal-phosphonate thin films include 
magnetic switches, in which the magnetic behavior of the inorganic lattice can be 
altered by a structural change in the organic region. Also, films containing a 
conductive or non-linear optic organic region as well as a magnetic inorganic lattice 
could act as a sensor. This dissertation will focus on the preparation of porphyrin 
containing zirconium phosphonate films where the metal phosphonate lattice acts to 
stabilize the films toward potential catalytic reaction conditions. 



24 



1 .3 Background on Porphyrins 

1.3.1 Optical Behavior of Porphyrins 

Porphyrins are a common research focus in physics, chemistry, and biology. 
Physical and chemical interest in porphyrins stems, for example, from their highly 
conjugated structure that allows facile electron-transfer,68-72 and from their chemical 
actiyity at an exposed metal that may be active toward catalysis or chemical 
sensing. 1.8,73-75 Biologists and biochemists are interested in the common biological 
building blocks that are based on the porphyrin structure. 76-78 

The core structure of the porphyrin is the completely saturated porphine 
macrocycle (Figure 1.1 1).79 Upon reducing this macrocycle to the unsaturated form, 
the porphyrin chromophore is achieved. By hydrolyzing one of the pyrrole units, the 
chlorin compound is prepared. Another important structure based on the porphine 
core is the phthalocyanine or the tetraazatetrabenzporphyrin. Each of these structures 
includes either two protons, as in the free base porphyrin, or a coordinated metal 
within the center of the porphyrin, called the metallo-porphyrin. Examples of 
biologically active porphyrins include chlorophyll, which is a manganese-coordinated 
chlorin molecule, and heme, which is an iron-substituted porphyrin. 



25 




B 



Figure 1.11: Structures of poiphyrin-type molecules A) porphine B) free base 
porphyrin, and C) pthalocyanine. 



Porphyrins have characteristic and strong optical transitions by which they can 
be identified. The bands often observed in visible spectra of porphyrins include the B 
or Soret Band and Q Bands, as seen in Figure 1.12 for a palladium 
tetraphenylporphyrin (PdTPP). The Soret Band is associated with the allowed n-n* 
transition and is typically seen between 380 and 420 nm.^o 




400 



500 



600 



Wavelength (nm) 



Figure 1.12: UV-vis spectrum of a metallo-porphyrin (PdTPP). 



M 



The Q-Bands are observed between 500 and 600 nm. The lower energy Q- 
Band (Q„) is associated with the electronic origin, Q(0,0) of the lower energy singlet 
excited state. The higher energy Q-Band (Qp) has a contribution from a vibration 
mode and is denoted Q(1,0). Both Q(0,0) and Q(1,0) are quasi-allowed transitions 
with relatively low absorbance intentisties. The Q-Bands are highly sensitive to the 
symmetry of the molecule. In porphyrins of 04^ symmetry such as metalloporphyrins, 
or the diacidic or dibasic forms of the porphyrin, two Q Bands are observed as 
pictured in Figure 1.12. The free-base porphyrin is of D2h symmetry and the 
degeneracy of the Q-Bands is disrupted, splitting the Q-Bands into four peaks.^o 

The above described transitions are due to the porphyrin Ti-electrons and are 
7i-7t* in nature. If these transitions are unperturbed by the central substituent, the 




i 


\ 




I 




i 

Q 


I 


B 



Figure 1.13: Outline of 1 6-member principal resonance structure of metallo- 
porphyrin. 



porphyrin is classified as "regular". Similarly, the emission spectra of regular 
porphyrins are determined solely by the chromophore itself The above explanation of 
the UV-visible behavior of porphyrins is based on the free-electron model, in which 
the core of the porphyrin, the 1 6-member heterocyclic, conjugated ring behaves like a 
free-electron wire (Figure 1.13). Another popular theory is Gouterman's four-orbital 



27 



^-■> 



it 



model (Figure 1.14), which combines the Huckel-MO theory with the free-electron 
model. In this model, Gouterman describes four orbitals, two LUMOs, c,(e ) and C2 
(Cg) each with five nodes, which are degenerate in energy, and two HOMOs, b,(a2j 
and b2(a,J, each with four nodes, which are not degenerate. According to the four 
orbital model, the Soret Band corresponds to the transition from the lower energy a,„ 
orbital to the e^ orbital, giving a higher energy transition. The Q-Bands arise from the 
transition from the a2u orbital, which is higher in energy giving a lower energy 
transition.25,80 




Ci(ea), 

C2(eq) 



i\ i\. 



bi(a2u> 



Maiut 



Figure 1.14: Gouterman's four-orbital model. 



28 



However, there are also "irregular" porphyrins. Irregular porphyrins, typically 
metalloporphyrins, are broken down into categories called hypso- and hyper- 
porphyrins. In the case of irregular porphyrins, the central metal contains partially 
filled shells, which introduce a possibility of metal electrons mixing with porphyrin n- 
electrons. This mixing is caused by the possibility of metal to porphyrin back-binding 
due to similar energies of the metals d-orbitals and the porphyrin's 7t-orbitals. The 
central metal ion can lead to significant changes in the optical and emission spectra. 
The metal and its oxidation state determine which category the porphyrin's optical 
behavior will fall into. Also, the release of electron density from the metal to the 
porphyrin enables the metal to stay co-planar with the chromophore as the effective 
size of the metal is reduced.'^^ 

Hypsoporphyrins have central metals of groups eight through eleven with 
configurations d"" where m = 6 - 9 and have filled eg(d7i) orbitals. The inclusion of 
these metal ions is often associated with a bathochromic or blue shift relative to the 
corresponding free base porphyrin. Common hypsoporphyrins include Ni(II), Pd(II), 
and Pt(II)-porphyrins. The Ni(II) porphyrins are easily affected by basic axial ligands, 
whereas Pd(II) and Pt(II) are typically four coordinate and appear insensitive to the 
potential ligand envirormient ^0-8' 

The second class of irregular porphyrins is called the hyperporphyrins, which 
is fiirther broken down into subclasses called p-type, d-type, and pseudonormal 
hyperporphyrins. Most metallo-porphyrins classified as hyperporphyrins have central 
metals with easily accessible lower oxidafion states. Of these, Mn(III) and Fe(III) are 
the most well studied due to their biological implications. The spectra of 
hyperporphyrins exhibit the Soret and Q-Bands as before with some possible shifting. 
Additional prominent absorption bands may be seen typically at higher energies 
relative to the Soret Band. The hyperporphyrin spectra demonstrate the effects due to 



-B 



metal-ligand charge transfer (MLCT) mixed with the porphyrin ti-tt* transitions even 
within the Soret Band. The MLCT Bands can be porphyrin to metal, metal to 
porphyrin, or even axial ligand to metal. Due to the spectral sensitivity to 
chromophore substituents, to the metal its oxidation state, and to the nature of the axial 
ligand, and the additional UV-vis Bands, hyperporphyrin spectra are much more 
difficult to analyze.^0,81 

The d-type hyperporphyrins include metals of groups six through eight. 
Mn(III)-porphyrin, for example, is d" and S = 2, or high-spin, and is a characteristic d- 
type porphyrin. In chloroform, Mn(IIl) tetraphenylporphyrin shows six peaks. An 
early researcher of the Mn-porphyrins, Boucher, termed these peaks by Roman 
numerals going from low to high energy. The first two peaks are in the far-red region 
between 800 and 650 nm. Bands III and IV absorb in a region similar to the Q-Bands 
in regular porphyrins, between 500 and 650 nm. Band V is similar to and often called 
the Soret Band though this band now includes contributions from metal to ligand 
mixing. Band VI is typically around 350 nm. The ratio of Bands V and VI is very 
sensitive to axial ligands and ring substituents. These bands are due to porphyrin to 
metal charge transfer a,„(7r), a2„(7i) to qJ^&k), which implies a necessity for one or more 
vacancies in the Qj^dn) orbital of the metal and reduction potentials which are not too 
negative.80'8' 

Finally, pseudonormal hyperporphyrins include VO(IV), Cr(II), Mn(II), 
Mo(IV), La and Ac where S ^t 0. These metals show normal absorption spectra with a 
weak extra absorption possible in the far-red region. All of these metals have a 
partially filled or empty eg(d7i) orbital, but charge-transfers from the porphyrin to the 
metal are too high in energy to be observed in the UV-vis region. In addition, fiirther 
reduction takes these metals to unstable oxidation states which makes this an even 
higher energy transition and highly unlikely 80,81 



30 



In addition to intramolecular effects such as the metal, substituents, and axial 
ligands, intermolecular effects such as aggregation can significantly alter the electronic 
behavior of porphyrins. Aggregation in these chromophores has been described 
thoroughly by Kasha's exciton theory. This theory looks at aggregation only from the 
point of view of overlapping transition dipole moments (Figure 1.15) and not as 
interacting 7t-systems. In metallo-porphyrins, the transition dipole moments are 
equivalent due to the symmetry of the chromophore.'^6'82.83 




— M. 



Figure 1.15: Transition dipole moments in metallo-porphyrin. 

Aggregation, according to Kasha's model, splits the excitation energy of the 
monomer (E") into high and low energy components. Equation 1.8 describes the 
energy dependence on aggregation by: 



E" =E° +D±V 



(1.8) 



where D is the dispersion energy which is highly dependent on the change in 
environment upon aggregation, and V is the exciton splitting energy. 76,83,84 



' V 31 



3i 



When the chromophores are interacting with the transition dipole moments 
parallel, the exciton energy can be described by Equation 1 .9: 

where M is the transition dipole moment, R is the center-to-center distance, N is the 
number of chromophores, and a is the angle between R and M (Figure 1.16). So, if a 
< 54.7°, V will be positive, and the exciton splitting will be greater and a red shift will 
be observed as the transition shifts to lower energy. A red shift is observed in what are 
called J-aggregates where both M^ and M^ make angles less than 54.7° with the R 
vector. If a > 54.7°, V will be negative, and the exciton splitting energy will be lower 
leading to a blue shift in the spectrum. When M^ and My are both greater than 54.7° 
fi-om R, the aggregates are termed H-type. If a, < 54.7° and a^ > 54.7°, the spectral 
components will split and part of the band will shift red and part will shift blue; this 
spectral behavior is seen in edge-to-edge type aggregates. There can be combinations 
and varying degrees of these types of interactions within an aggregated domain 
possibly leading to complicated spectra, but in general, the optical spectra ease 
identification of electronic behavior of porphyrin chromophores (Figure 1.16). 76,83,84 



32 



S. 



E. 




















B 



Figure 1.16: Porphyrin chromophore interactions: The square represents the 
chromophore and its disecting axes. A) H-type or face-to-face aggregates; B) edge-to- 
edge aggregates; C) J-type or head-to-tail aggregates. 



The transition dipoles M^ and M^, are typically parallel to the plane of the 
chromophore except when the nature of the metal in the chromophore center causes a 
puckering of the ring. Therefore, polarized UV-vis experiments can easily indicate 
orientational changes of the chromophore within a film (Figure 1.16).''6,83,84 

Incorporation of porphyrins into LB films is currently of interest in scientific 
literature. These films are designed in order to prepare selective gas-sensors, ''"^^js 
photovoltaic devices,^^ electron-transfer materials,''2,86 molecular wires,^"^ and novel 
heterogeneous catalyst systems.^^ However, a difficulty arises in the stability of the 
samples using the "typical" LB methods of purely hydrophilic/hydrophobic 
interactions or by self-assembly involving tethering through a ligand. Including a 
metal-phosphonate lattice into these films should significantly improve the stability 
and the applicability of these materials in ultra-thin, organized films. 

Typically, LB films containing porphyrin constituents have been studied in 
which the chromophore itself is the polar head group. These porphyrin films have 
been successfiiUy prepared with either the molecule sufficiently diluted with a fihn 
stabilizing amphiphile,33,88-90 such as stearic acid, or with long hydrophobic chains 



33 



attached to the chromophore to stabiUze the monolayer on the water surface.^'*'^' 
There are significant disadvantages to this method of film preparation. First, the 
chromophore is buried in the film interior on a transfer onto a hydrophilic substrate, 
and commonly, the hydrophobic interactions necessary to deposit onto a hydrophobic 
substrate are too weak for successful transfer. Also, the hydrophilic interactions are 
typically of a hydrogen-binding nature making this a relatively weak interaction 
destabilizing the film. Finally, the conditions necessary for transferring the traditional 
porphyrin LB films facilitate aggregate formation, which can be detrimental in certain 
applications, such as catalysis. 

The aggregation, or chromophore n-n interactions, is often a consequence of 
the film forming procedures. First, compression of the film on the water surface 
forces the eventual overlap or tilting of the chromophores-^'^'^^-^o Also, the decreased 
affinity of the derivatized chromophores for water tends to force the chromophores to 
aggregate rather than to spread on the water surface.92 Understanding the molecular 
orientation, aggregation, and morphology of porphyrin LB films is critical because 
each is intimately linked to chromophore behavior. For example, aggregation can 
significantly reduce or eliminate the efficiency of the porphyrin in catalysis''^ or the 
ability of the porphyrin to bind probe molecules in a sensor.^^ Therefore, it is 
desirable to find methods for forming porphyrin LB films with no aggregation. 

1 .3.2. Background on Manganese Porphyrins 

Biomimetic systems involving porphyrin catalysts have often been discussed in 
scientific literature over the past 20 years. Manganese and iron porphyrins are 
commonly studied oxidation catalysts and are prevalent elements in biological 
processes.93-96 Biochemical oxidation reactions employing metallo-porphyrins 
involve reversible site-specific binding of the substrate such that the substrate is within 



34 



reach of the oxygen atom on the metal. After the oxygen has been successftilly 
transferred to the substrate, the product is released and the catalyst is regenerated.^^ 

Manganese porphyrins are probably most well known as epoxidation and 
hydroxylation catalysts whether under heterogeneous or homogeneous conditions. 
The manganese porphyrin catalysts can utilize a number of different oxidants such as 
iodosylarenes, alkylhydroperoxides, hydrogen peroxides, and perchlorates among 
others, in order to accomplish the facile oxidation of deactivated olefins, alkanes, 
alcohols, ethers, and amines.^'^^'O^ a hyper- valent metal-oxo species is believed to be 
the active intermediate in the oxidation process in cases such as dioxygen activation of 
Cytochrome P-450, or in oxygen transfer from iodosylbenzene, peracids, or 
hypochlorite oxidants. '^i Though there is some debate on the actual mechanism of the 
epoxidation, there are a few possible routes (Figure 1.17). The suggested first and 
rate-determining step is the formation of a charge-transfer complex. Whether the 
reaction then proceeds through epoxidation or rearrangement is dependent on the 
oxidation potentials of the alkenes and the oxidants, steric and electronic structures of 
the reactants, and the ability of the substrates to undergo rearrangement.^'^ 

Porphyrins have also been studied in chiral catalysis. Lai and co-workers studied 
the asymmetric aziridation of alkenes using a chiral manganese porphyrin catalyst. '02 
They found that with bulky chiral substituents on the porphyrin, successful nitrene 
transfer to alkenes was achieved. Enantiomeric excess ranging from 43 to 68% and 
product yields greater than 70% were obtained. In these catalysis studies, the reactive 
intermediate was a Mn(IV) complex. 



35 



vV 




concerted 
oxene insertion 






Cm^ 



oxidant 




/=\ 



/\ 



possible rate-limiting 
formation of a charge- 
transfer complex 

Figure 1.17: Suggested mechanism of olefm epoxidation catalyzed by MnTPP, 



1.3.3. Immobilization of Porphyrins 

The ability of porphyrins to efficiently catalyze both the epoxidation of olefins 
and the hydroxylation of alkanes unfortunately leaves the porphyrin and its 
superstructure vulnerable as potential substrates. However, nature has developed 
mechanisms to eliminate these unwanted complications. For example, an enzyme and 
its cofactors may form metal-oxo complexes only when the substrate molecule is 
confined within an enzymatic cavity. Also, the tertiary protein structure prevents the 
active porphyrin catalyst from approaching other potentially oxidizable 
metalloporphyrins, and it makes the structure rigid, protecting the amino acid 
backbone and the side-chains from intermolecular oxidation by contacting the active 
site. These biosystems are difficult to mimic in the laboratory; however, successful 



36 



biomimetic catalysts have been prepared with bulky, rigid groups substituted on the 
porphyrin chromophore.'^i 

One alternative solution to the problem of internal oxidation or intermolecular 
oxidative destruction of the porphyrin catalyst is immobilization of the chromophore. 
Immobilization involves tethering the porphyrin to a surface such as a film,88 an 
inorganic solid particle, '03-106 a polymer, '07,1 08 a membrane, '09 or a resin. "O 
Immobilized porphyrins as biomimetics and as heterogeneous catalysts have been well 
explored in the past several years. ' 04- 1 06, i lo Tethering of porphyrins to a solid support 
can not only reduce or eliminate oxidative destruction of the active catalyst, but can 
also aid in the catalyst recovery after the reaction is completed. 

Heterocyclic ligands are commonly used as the link between the porphyrin and 
surface in many immobilized porphyrin systems.94,98,11 1 Unfortunately, binding the 
metallo-porphyrin to the imidazole allows little control over the porphyrin orientation 
in the films. Additionally, in these circumstances, there is no chemical connection 
between the porphyrin and the surface other than the ligand, which leaves the 
porphyrin vulnerable to removal from the surface by ligand displacement, changing 
the reaction conditions.30,94,98,1 1 1 ^n alternative method for tethering the porphyrins 
to surfaces has been established, which uses four alkyl phosphonic acid substituents 
that can be attached to a zirconium phosphonate network making a very stable 
catalytic film. 

1.3.4 Heterocyclic Ligand Cocatalvsts 

1.3.4.1. Li gand activation of the porphyrin catalvst . Heterocyclic ligands are 
well documented in the literature as activating Fe(III) and Mn(III) porphyrins for 
catalysis with oxidants such as alkyl or hydrogen peroxides.94,98,99 Porphyrins 



37 



immobilized on an ion-exchange resin support showed significant increases in 
catalytic activity in the presence of either imidazole or 4-methylimidazole. With the 
heterocyclic ligand present, nearly quantitative conversion of cyclooctene to 
cyclooctene oxide was achieved, relative to only 5% conversion in the absence of 
imidazole over the same time period. "2 Likewdse, Arasasingham et al. found a 4 to 
10 fold increase in the rate of the reaction between a manganese porphyrin and an 
oxygen source commonly used in olefin epoxidation reactions, t-BuOOH, in the 
presence of imidazole. Since the oxidation of the porphyrin accelerates, a rate increase 
should also be observed in the overall epoxidation reaction.98 

According to Yuan and Bruice, the reaction of the Mn(III)TPP CI complexes 
with peroxide oxidants only proceeds in the presence of a heterocyclic nitrogen base 
ligand such as imidazole or pyridine. The imidazole ligation was pH dependent and 
was evident only above pH 5. Consequently, the enhanced oxidation rate was also pH 
dependent. Further, with common oxidants, nitrogen base ligation led to a significant 
increase in the oxygen transfer rate. ^ ' • 

The rate increase could be due to a general-base catalysis and/or ligation of the 
imidazole (ImH) to the manganese ion.^s Activation by ligation of ImH is supported 
by the fact that when 2,4,6-trimethyl-pyridine is used as the base, which is sterically 
forbidden from porphyrin ligation, no increase in the reaction rate was observed. 
However, when the ImH concentration was below a saturation level, the rate increase 
was linear with ImH concentration up to a saturation level. An increase in the 
oxidation rate with a basic ligand is likely due to the increase in the electron density at 
the metal center arising from donation of the lone pair of electrons from the ImH.^s 
The presence of the ImH as an axial ligand has been shown also to stabilize the metal- 
0X0 compound. ^9 



38 



1 ■3.4.2. Spe ctral evidence of axial lip and. ImH to porphyrin binding should be 
apparent in the UV-vis spectra. The formation of a bis-imidazole Mn(III)TPP CI 
complex was demonstrated by a broadening shift in the Soret Band from 478 nm for 
the pure porphyrin to 472 nm. ' 1 1 Interestingly, the equilibrium constants for the 
formation of the mono- and bis-ligated imidazole-porphyrin complexes are similar and 
their absorption spectra are nearly identical. Therefore, at high concentrations of 
imidazole in solution, it is possible that the observed spectra arise from the formation 
of bis-imidazole complexes." !'• '^ However, the preferred formation of the mono- vs. 
bis-imidazole complexes has caused some disagreement, and some authors claim that 
even at saturated concentrations of imidazole, the principal component is the mono- 
imidazoleonly."4 

In the UV-vis of the Mn-porphyrin, the Soret Band of the Mn-porphyrins is the 
most sensitive to the axial ligand. The change in the Soret energy is due to the charge 
induced on the porphyrin chromophore through the metal. Electron-donating axial 
ligands induce negative charge on the macrocycle, separating the bonding and anti- 
bonding orbitals of the porphyrin, and increasing the transition energy.! 15 Hard 
anions, whose binding is strengthened by increased ionic character of the central 
metal, prefer localization of positive charge on the metal, which leaves the 
chromophore with more negative charge.' 15 Similarly, as a basic ligand takes on more 
hard base character, the A.^3^ will shift to higher energies. 

1 .4 Dissertation Overview 

The overall goal of this dissertation was to prepare zirconium phosphonate thin 
films by both the SA and LB technique that contained catalytic Mn-porphyrins. The 
purposes of the zirconium phosphonate network were to stablize the manganese 



\> '"-■■^■f 39 



containing films to reaction conditions and to allow these films to be recycled in a 
number of catalytic studies. Chapter 2 is an overview of the experimental techniques 
used to prepare and characterize the films described in this dissertation, and materials 
and instrumentation used in this pursuit are also presented. 

Films containing a Pd-tetraphenyl porphyrin were prepared to develop fihn 
preparation procedures and to better analyze the UV-vis properties of porphyrin 
containing films. Substituted tetraphenyl porphyrins, palladium 5,10,15,20- 
tetrakis(2,3,5,6-tetrafluorophenyl-4-octadecyloxyphosphonicacid)porphyrin(PdP4) 
and palladium 5,10,15-tris(2,6-dichlorophenyl)-20- (2,3,5,6-tetrafluorophenyl-4- 
octadecyloxyphosphonic acid)porphyrin (PdPl), have been studied as Langmuir 
monolayers and as zirconium phosphonate LB and SA films. 

Films were prepared incorporating the pure porphyrins and the porphyrins 
mixed with octadecylphosphonic acid (ODPA). The Langmuir monolayers were 
characterized with pressure vs. area isotherms and reflectance UV-vis spectroscopy. 
Using a three-step deposition technique, symmetric and alternating zirconium 
phosphonate bilayers and multilayers were prepared by the LB technique. PdP4 
containing films were also prepared by the SA technique. In all PdPl and PdP4 films, 
the porphyrin constituent resided in the hydrophobic region of the monolayer and the 
phosphonate substituents bound zirconium ions in the hydrophilic region. 

LB and SA films were studied with transmittance UV-vis and the LB films 
were fiirther investigated using X-ray diffraction. Control over chromophore 
interaction was achieved by chemical modification of the amphiphiles and by selection 
of appropriate transfer conditions. For example, reduced aggregation was seen in LB 
films of the tetraphosphonic acid substituted porphyrin PdP4 transferred at mean 
molecular areas (MMA) larger than the area per molecule of the substituted porphyrin 
and m SA films. In these films, the porphyrin macrocycles are non-aggregated and 



40 



oriented parallel to the surface. In contrast, the monophosphonic acid substituted 
PdPl aggregates under all of the deposition conditions studied. 

Stability of the Pd-porphyrin LB and SA films was examined by exposing the 
films to refluxing chloroform. UV-vis absorbance after immersion in chloroform 
confirmed conclusions that in films of PdPl, many of the chromophores are not 
tethered to the inorganic network and are easily removed, whereas in films of PdP4, all 
molecules bind to the zirconium phosphonate extended network making these films 
very resilient. 

Study of the Pd-porphyrins led to significant understanding of the behavior of 
tetra- and mono-phosphonic acid porphyrins in LB films. Chapter 3 describes the 
results of these studies, which were the first to show incorporation of porphyrins at the 
exterior of metal-lattice containing films. 

Manganese tetraphenyl porphyrins are well known epoxidation 
catalysts,^'''99'"6 and the incorporation of these catalysts into zirconium phosphonate 
films should improve their catalytic efficiency as well as their stability and 
recoverability. Films containing manganese 5,10,15,20-tetrakis(2,3,5,6- 
tetrafluorophenyl-4-octadecyloxyphosphonic acid)porphyrin (MnP4) have been 
prepared using the LB and SA techniques. The formation of these films involved 
modifying traditional LB procedures with SA techniques, which is possible with the 
use of zirconium phosphonate networks. From Langmuir monolayer and LB studies 
of the pure tetraphosphonic acid porphyrin, it appears that the MnP4 amphiphiles tend 
to form face-to-face aggregates, or H-aggregates, when assembled at the air-water 
interface, and this aggregation is translated into the transferred films. Attenuated total 
reflectance (ATR) IR, UV-vis, XPS and stability studies confirm the presence of the 
porphyrin. Thorough characterization of the MnP4 containing films is described in 
Chapter 4. 



41 



The heterocyclic imidazole ligand has been shown to improve the catalytic 
efficiency of Mn-porphyrins, and MnP4 films containing the imidazole ligand have 
been successfully developed. These films were prepared by a variety of methods 
involving a combination of LB, SA and substitution procedures. In solution, it is seen 
that binding of a non-amphiphilic imidazole causes a small blue shift of the Mn- 
porphyrin Soret band; however, a dominant influence on the Soret band in the films 
and in solutions containing the ImODPA ligand comes from the metals axial 
environment ~ especially halide binding. Mixed films containing both the imidazole 
phosphonic acid (ImODPA) and the MnP4 molecules have been prepared and 
characterized by ATR-IR, UV-vis, and XPS. The preparation and characterization of 
imidazole and MnP4 containing films is presented in Chapter 5. 

The epoxidation of cyclooctene using iodosylbenzene was catalyzed by the 
pure MnP4 containing films with substrate to oxidant ratios of 20:5, 40:5, and 60:5 
over a variety of reaction times. The self-assembled MnP4 films proved to have 
slightly improved catalytic efficiency relative to the analogous LB films likely due to 
the increased aggregation observed in LB deposited films. The mixed 
ImODPA/MnP4 films showed catalytic activity in the presence of the peroxide 
oxidants. These films were examined with different substrate to oxidant ratios. The 
porphyrin containing films, both with and without ImODPA were resistant to 
degradation under most examined reaction conditions. The catalysis results involving 
both PhIO and H2O2 oxidants with pure porphyrin and mixed porphyrin/imidazole 
films are described in Chapter 6. 



CHAPTER 2 
EXPERIMENTAL 

2.1 Langmuir-Blodgett and Self-Assembled Films 

2.1.1. General Langmuir-Blodgett and Self- Assembly Procedures 

2.1.1.1. Fil m Formation. The general procedure for forming LB films starts 
with the Langmuir monolayer, which are prepared on a Langmuir trough. The trough 
consists of a rectangular piece of Teflon, typically 1 cm deep, supported on a metal 
base with Teflon barriers, shown as black rectangles in Figure 2.1. A Teflon well is 
carved in the center of a double barrier trough for transferring monolayers. The 
spreading solution is prepared by dissolving the amphiphile of interest in a volatile 
solvent, such as CHCI3. The solution is carefully applied to the subphase surface, 
ideally spreading the molecules uniformly over the surface. 



Barrier 



Aqueous 
Subphase 



Wilhelmy 
Balance 




Syringe 



Surfactant 
molecule 



Teflon Trough 



Wilhelmy 
Plate 



Figure 2.1: Schematic of Langmuir-Blodgett trough and monolayer. 



42 



' -■: :^ 43 

The amphiphiles are shown in Figure 2.1 as gray circles, representing the 
hydrophilic head group, and black lines, representing the hydrophobic tails. The 
subphase, which is usually aqueous, must be nanopure. Because there is such a small 
amount of amphiphile present, the monolayer is extremely sensitive to contaminants - 
especially lipids and other surfactants and ions found in soaps and tap water. 

The barriers compress the amphiphiles at a constant speed. In studying the U- 
A isotherm, the film is compressed until it collapses. For LB transfers, the film is 
compressed until the desired transfer pressure is achieved. At this point, the 
monolayer is held at constant pressure for approximately two minutes until the 
monolayer is stabilized, then the solid substrate is dipped vertically down through this 
monolayer. 

The monolayers are first characterized with TI-A isotherms. In modem 
computer operated systems, the concentration (mg mL"') and the molecular weight (g 
mol'), or the concentration in mol L"', of the compound being spread is entered into 
the program along with the spreading surface area in mml From this information, the 
program can calculate the MMA in A^ molecule "'. As the barriers move together and 
the surface is compressed, the effective MMA is decreased and the surface pressure 
increases. 

The preparation of the zirconium phosphonate porphyrin films took place by a 
three-step deposition procedure (Figure 2.2).28,29,38 a glass sample vial was placed in 
the subphase in the well of the trough. Octadecylphosphonic acid was spread from 0.3 
mg mL-' CHCI3 solutions and compressed at 15 - 20 mm min-' on the water surface. 
At 20 mN m-', the substrate was dipped down through the monolayer surface and into 
the sample vial at 8 mm min-i, transferring the ODPA template layer. The substrate 
and the vial were then removed from the trough and an amount of zirconyl chloride 
was added to the vial to make the solution ca. 4 x 10-5 M in Zr4+. After 20 min in the 



44 



zirconium solution, the substrate was removed from the vial and rinsed with water. 
After the template layer was successfully prepared, it was dried and characterized 
independently by ATR-IR, XPS, and UV-vis if needed. 

Capping layers were prepared by a variety of methods, which will be described 
for each different film type in Chapters 3, 4 and 5. LB or SA methods could be used 
to form the capping layers. To form the capping layer and complete the zirconium 
phosphonate bilayer by the LB technique, the now hydrophilic substrate was lowered 
into the trough, a monolayer was spread on the surface and compressed to the desired 
pressure, and the substrate was raised through the monolayer at 5 mm min'. To form 
the capping layer by self-assembly, the hydrophilic surface was submerged in a 
solution of the desired molecules at about 1 0"' M in an appropriate solvent, usually 
EtOH/H20 (9/1). The capping layer was then allowed to self-assemble. 

2.1.1 .2. Materials and Methods . Materials used to prepare the porphyrin 
containing films included octadecylphosphonic acid (ODPA), zirconyl chloride 
(ZnOCl-8H20), and the porphyrins themselves. The porphyrins were provided by 
Bruno Bujoli, Fabrice Odobel, Karine LeClair, and Laurent Camus fi-om the 
Laboratoire de Synthese Organique, at the Faculte des Sciences et des Techniques de 
Nantes in Nantes, France. ODPA was used as purchased from Alfa Aesar (Ward Hill, 
MA). Zirconyl chloride, 98% was used as supplied from Aldrich (Milwaukee, WI). 
Octadecyltrichlorosilane (OTS) 95%, used to silanize and hence hydrophobicize the 
substrates, was also used as purchased fi-om Aldrich. Amylene stabilized HPLC grade 
CHCI3 was used as a spreading solvent, and was used as received from Acros 
(Pittsburgh, PA) and Fisher Scientific (Pittsburgh, PA). 

A KSV 2000 system (Stratford, CT) was used in combination with a 
homemade, double barrier Teflon trough for the Langmuir monolayer studies and LB 
film preparation. 



45 



STEP1 



'■■i- » 



STEP 2 



i 



Transfer template 
from water surface 




»— 


— « 


■- 




* — 


— <j 






V — 


— « 






g -- 




4 

.J 




f" 







Sample vial in trough 



Zr^"(*; 



•o 


— »« 


« -^ 














- -; • 


w . 


J • 


w 








w 








w ' 




























<;^____ 


'Z^ 



STEP 3 



SA Porphyrin 



»-«•« — 


», ♦-« 




1 — »•« 


-— «»<^H 




































,1 






^ ! 








i 


^~~II__ 








1 



LB 



•• — — ♦ 

♦ — — • 
0» — — • 
«• — — • 

• * • 



•- 



III r/::= =:•: .^ I II 






Figure 2.2: Schematic of the three-step deposition process used for zirconium 
phosphonate films. 



46 



The surface area of the 2000 trough was 343 cm^ (36.5 cm x 9.4 cm). A 
platinum or filter paper Wilhelmy plate, suspended from a KSV microbalance, 
measured the surface pressure. Subphases were usually pure water with a resistivity of 
17-18 MQ cm-i produced from a Bamstead NANOpure (Boston, MA) purification 
system. 

The films were transferred from the aqueous surface onto solid supports. Glass 
microscope slides and glass coverslips were purchased from Fischer (Pittsburgh, PA) 
and were used as substrates for UV-vis and catalysis studies. Single crystal silicon 
wafers (10 0) were purchased from Semiconductor Processing Company (Boston, 
MA), and cut using a diamond glass cutter to 25 mm x 15 mm x 0.8 mm for XPS 
studies. These substrates were cleaned using piranha etch, which is 1 :4 H2SO4: 30% 
H2O2, a new hydrophilic surface was prepared by the RCA procedure, • 17 which 
involved first, heating in a 5:1:1 solution of water, 30% H2O2, and NH4OH, and 
second, heating in a 6: 1 : 1 solution of water, 30% H2O2 and HCl. Then the substrates 
were sonicated for 15 minutes each in methanol, 50/50 by volume methanol/ 
chloroform, and chloroform. The substrates were then sonicated in a 2% 
octadecyltrichlorosilane (OTS) solution in hexadecane and CHCI3 for two hours. 
Finally, the substrates were sonicated for 15 minutes each in CHCI3, 50/50 by volume 
CH3OH/ CHCI3, and CHjOH.'is 

2.1.2. Characterization 

2.1.2.1. UV-vis. Transmittance UV-visible experiments were performed on a 
Cary 50 spectrophotometer by Varian with an average resolution of 2 nm. Porphyrin 
solutions were studied by UV-vis in EtOH, H2O, CHCI3, and CH2CI2 solvents. The 
behavior of the porphyrin with different potential ligands was investigated by mixing 
the porphyrin solution with ethylphosphonic acid, t-butyl ammonium halides (chloride 



47 



and bromide) (Aldrich), and imidazole with no alkyl substituents (ImH) (Kodak). A 1 
cm X 1 cm X 3 cm quartz cuvette held the sample, and the background using the 
corresponding pure solvent was subtracted. 

A Teflon substrate holder with grooves cut at 45° to one another was used to 
obtain sampling at 0° (beam normal to the substrate) and 45° incidence to the substrate 
surface. A plane, visible polarizer was used to select s- and p-polarized light. 
Reflectance UV-vis experiments were performed on a KSV 2000 mini-trough using an 
Oriel spectrophotometer and a 77410 filter with a range from 200 - 600 nm. 

2.1.2.2. X-ray Photoelectron Spectroscopv. X-ray photoelectron spectroscopy 
(XPS) was performed on a Perkin-Elmer PHI 5000 Series spectrometer using the Mg 
Ka line source at 1253.6 eV. The instrumental resolution was 2.0 eV, with anode 
voltage and power settings of 15 kV and 300 W, respectively. The operating pressure 
was around 5x10'' atm. Survey scans were performed at a 45° takeoff angle with a 
pass energy of 89.45 eV. During multiplex scans, 80-100 scans were run at each 
peak over a 20-40 eV range with a pass energy of 37.35 eV. 

2.1 .2.3 X-rav Diffraction . In order to obtain low angle X-ray diffraction 
(XRD) patterns, multilayer films were transferred onto a hydrophobic glass slide. The 
diffraction patterns were obtained using a Phillips APD 3720 X-ray powder 
diffractometer with the CuKa line, A, = 1.54 A, as the source for films ranging from 10 
to 15 bilayers. 

2.1.2.4. Attenuated Total Reflectance Infrared . Attenuated total reflectance 
infrared spectroscopy was performed on a Mattson Instruments (Madison, WI) 
Research Series- 1 FTIR spectrometer equipped with a deuterated triglyceride sulfide 
detector and a Harrick (Ossining, NY) TMP stage which held the Ge crystal substrate. 
ATR-FTIR spectra consisted of 500 scans at 2 cm-' resolufion and were referenced to 
the silanized crystal or previous bilayers. 



': \( ' 



■■;<^ 



48 



2.2 Porphyrin Films 

2.2.1. Palladium Porphyrin Films 

The palladium porphyrins studied were palladium 5,10,15,20-tetrakis(2,3,5,6- 
tetrafluorophenyl-4-octadecyloxyphosphonic acid)porphyrin (PdP4) and palladium 
5, 1 0, 1 5-tris(2,6-dichlorophenyl)-20- (2,3 ,5,6-tetrafluorophenyl-4- 
octadecyloxyphosphonic acid)porphyrin (PdPl). The Bujoli group provided us with 
these porphyrin amphiphiles. 

The Pd-porphyrins made well-behaved monolayers; therefore, these 
amphiphiles were often transferred by the LB technique. Additionally, the Pd- 
porphyrins were studied in diluted mixtures with ODPA in an attempt to disrupt the 
aggregate formation in the films. For mixed, Pd-porphyrin/ODPA films, the two 
materials were simultaneously dissolved in a CHCI3 solution. The weighted average 
concentration and molecular weight were calculated and used in the KSV software to 
monitor the MMA with compression. Ratios of PdP (1 and 4) to ODPA studied 
included 1:0, 1:1, 1:4, 1:9 and 0:1, respectively. 

The creep of the pure palladium-porphyrin Langmuir monolayers was studied 
at high and low pressures over 30 min, or slightly longer than the time of one 
deposition. At a constant pressure of 12 - 15 mN m-', the area changed by 6% and 
12% for PdP4 and PdPl, respectively. At low pressure (3 - 5 mN m-'), the change in 
area was 3% and 7% for PdP4 and PdPl, respectively. The instability in the 
monolayers led to a necessary correction in the transfer ratios. The corrected transfer 
ratios for the pure PdP4 were 1 .0 - 1 .4 at high pressures and 1 .0 - 1 . 1 at low pressures. 
For the pure PdPl , the corrected transfer ratios were 0.8 - 1 .0 and 0.9 - 1 .0 for high 



49 



pressure and low pressure transfers, respectively. The transfer ratios of the porphyrin 
films mixed with ODPA consistently showed uncorrected transfer ratios near unity. 

Monolayers of PdPl and PdP4 were also studied on both heated and basic 
subphases. To heat the trough, an Isotemp Refrigerating Circulating Model 900 
(Fisher Scientific) pump with a water/ethylene glycol bath was used. The temperature 
of the subphase was monitored using a KSV thermosensor. A 0.01 M KOH solution 
was added to the subphase to adjust the pH to the desired value. 

PdP4 films were also studied by self-assembly. The SA solution was prepared 
by diluting 1 mL of a 0.5 mg mL' solution of PdP4 in CHCI3 to 30 mL with a 9/1 
EtOH/H20 mixture. The film was allowed to SA for approximately 2 hours before 
studying by UV-vis. 

2.2.2. Manganese Porphyrin Films 

Manganese 5,10,15,20-tetrakis(2,3,5,6-tetrafluorophenyl-4- 
octadecyloxyphosphonic acid)porphyrin (MnP4) and a model porphyrin, manganese 
5,10,15,20-tetrakis (pentafluorophenyl)porphyrin (MnPO) were prepared by the Bujoli 
group. Again, ODPA was used for the template layers, zirconyl chloride was used to 
prepare the zirconium network, and CHCI3 was used as the spreading solvent. Also, /- 
butylammonium chloride (/-BuNHi^ Cr) (Aldrich) was used as a chloride source for 
SA deposited films. NaCl (Acros) was used as the chloride source for LB transferred 
films. 

To form the MnP4 monolayers, a 0.4 mg mL' solution was prepared in CHCI3 
(often, in order to dissolve the porphyrins, up to 5% ethanol was added and the 
solution was sonicated for about an hour). An appropriate volume of solution was 
spread on the aqueous surface in order to reach and hold the desired transfer pressure 



50 



throughout the deposition. A variety of surface pressures were used for transfer and 
will be described in more detail in Chapter 4. 

To prepare SA Mn-porphyrin films, 1 mL of a 0.5 mg mL"' MnP4 solution in 
EtOH was diluted to 30 mL with a 9/1 EtOH/HjO mixture in a 50 mL vial. 
Alternatively, a 0.5 mg mL"' MnP4 solution in CHCI3 was diluted with pure CHCI3 or 
CHjClj. The zirconated ODPA surface was exposed to the SA solution for 2 hr unless 
otherwise specified. After 2 hr, there were typically some physisorbed chromophores, 
which were rinsed off the surface using a hot solvent such as CHCI3 or CH3CN. 

As is discussed in Chapter 4, halogenated solvents were originally chosen as 
self-assembly solvents to eliminate possible ethoxide or water binding. However, the 
oxide coating on the exposed zirconated ODPA template is soluble in EtOH/HjO 
solvents making the zirconium available for binding the phosphonic acids of the 
porphyrin capping layer. From UV-vis studies, the films formed fi-om these different 
solvent systems appeared to behave similarly. Therefore, because the film formation 
mechanism fi-om EtOH/H20 was better understood, this solvent mixture was normally 
used. 

To induce chloride binding at the Mn-porphyrin' s axial position, a SA solution 
containing approximately 0.01 - 0.1 M /-BuNH3'' CI" along with the porphyrin in 
EtOH/HjO was prepared. The zirconated ODPA template was submerged in this 
solution for 2 hr. Alternatively, chloride ions were incorporated into the aqueous 
subphase used for LB transfer of the MnP4 monolayers using NaCl at 0.1 M and 
greater concentrations. 

From UV-vis results of the MnP4 and MnPO with ethylphosphonic acid, it 
appeared that the phosphonic acid might cause the Mn(III)-porphyrin to go through a 
spin state crossover fi-om high-spin to low-spin Mn(III). In order to examine this 
magnefic change, the Evan's NMR method was used. "9. 120 a solufion of 10% /- 



51 



butanol in CDCI3 was injected into a small capillary tube using a long syringe needle. 
The depth of the solution in the capillary tube reached approximately 2" . Two 
standard NMR tubes were then filled to approximately 1" with sample solution. The 
first was filled with pure MnPO (0.0106 g, 100 ^mol) dissolved in the 10% t- 
butanol/CDClj solution. The second was filled with MnPO (0.0106 g, 100 ^imol) and 
ethylphosphonic acid (0.220 g, 2000 ^mol). The reference solution in the capillary 
was inserted into the porphyrin containing NMR sample, and the magnetic 
susceptibility of the solute, Xg (cm^ g"') induced by the magnetic porphyrin was 
approximated by: 



3 A/ 
ATtjm 



X,-^^ (2.1) 



where Af is the diamagnetic fi-equency shift, f is the spectrometer fi-equency, and m is 
the mass of substance per mL of solution. Compared to literature values, the 
differences in the Xg of MnPO with and without ethylphosphonic acid did not 
correspond to a spin state change in the Mn(III). 

2.2.3. Manganese Porphyrin/Imidazole Mixed Films 

The general method used to prepare MnP4/imidazole films in this study 
involved the initial formation of a zirconated octadecylphosphonic acid (ODPA) 
template, as before. Onto this template, a film of either pure imidazole 
octadecylphosphonic acid (ImODPA), which was prepared by the Bujoli group, or a 
mixture of ImODPA and MnP4 could be formed. The zirconium phosphonate 
network provided a means for locking the porphyrin and the imidazole into the films, 
resulting in films that were stable toward the conditions used for the catalysis 






52 



reactions. Even under relatively harsh conditions such as elevated temperatures or 
rapid solvent flow, the porphyrins films appeared to be stable. Additionally, the 
zirconium phosphonate network made preparation of the MnP4-imidazole films 
possible by a wide variety of mechanisms. 

MnP4 and ImODPA could be incorporated into both LB and SA films. In 
order to accommodate ImODPA into LB films, two types of spreading solutions were 
used: 1) ImODPA mixed with a stabilizing agent such as hexadecylphosphonic acid 
(HDPA), or 2) ImODPA mixed with the MnP4 amphiphile. Alone, the ImODPA did 
not form sufficiently stable monolayers for transfer. The porphyrin, MnP4 was 
substituted into the films of pure ImODPA or ImODPA/HDPA from an EtOH/HzO 
solution. Alternatively, the MnP4 film was transferred by the LB technique and then 
the ImODPA was substituted into these films. *, 

First, in the case of the mixed 25% ImODP A/75% HDPA films, 200 ^iL of a 
solution with a weighted average concentration of 0.17 mg/mL and a weighted average 
molecular weight of 329.86 mg mmol"' was spread on the water surface and 
compressed to 12 mN m"'. The film was transferred at 5 mm min"' on the upstroke 
completing the zirconium network. Films made in this way were abbreviated 
ODPA/Zr/25% ImODPA. These films were then placed in a solufion of the MnP4 at 
ca.10-5 M in 9/1 EtOH/HjO and the porphyrin phosphonic acids were allowed to 
substitute into the defect or vacant sites in the film for 2 hr. Transmission UV-vis of 
these films confirms the ability to include the porphyrins by this method, and the 
resulting films were called ODPA/Zr/25% ImODPA, SA MnP4. 

Films containing both the MnP4 and the ImODPA transferred by the LB 
technique from a mixed monolayer were also prepared. A 70/30 mixture of 
MnP4/ImODPA, respectively, was dissolved in CHCI3 with a weighted concentration 
and MW of 0.289 mg/mL and 992.70, respectively. 175 ^L of this solution was 



53 



spread and transferred at 3 mm min"' on the upstroke forming the ODPA/Zr/30:70 
MnP4:ImODPA films. 

For the self-assembly of the imidazole onto a zirconated ODPA template, 2 mL 
of a 0.5 mg mL' solution of ImODPA in EtOH was dissolved in a 9/1 EtOH/HzO 
mixture in a 50 mL vial. The substrate containing the zirconated ODPA template was 
placed in the vial and the film was allowed to self-assemble for 2 hr. When the self- 
assembly procedure was complete, the film was rinsed with nanopure water and dried 
with forced air. The MnP4 molecules were allowed to substitute into these films as 
described in section 2.3.2. Similarly, mixed MnP4 and ImODPA solutions were 
prepared at a variety of ratios in EtOH/HjO, and the mixed monolayer was allowed to 
self-assemble for 2 hr. 

The ImODPA was deprotonated as mentioned in Chapter 5 by soaking the 
mixed film in an EtOH solution containing /-butyl amine. The /-butyl amine was used 
to deprotonate the imidazole within the films so the ligand would be available for 
binding the central manganese. The concentration of this solution was not precise but 
was consistently ca. 0.1 M and rinsing times were ca. 15 min. 

2.3 Catalysis 

2.3.1 ■ Catalvsis using PhIO as an oxidant 

Pure porphyrin films behaved as catalysts for the epoxidation of cyclooctene 
using iodosylbenzene (PhIO) as the oxidant. PhIO was synthesized fi-om the diacetate 
precursor using NaOH.i^i lodobenzene diacetate (3.0 g, 9.3 mmol) was placed in an 
Erlenmeyer flask. 30 mL of 3 NNaOH was added with stirring over 5 min. The 
mixture was stirred for 1 5 min and left to sit uncovered 45 min. 1 00 mL deionized 
H2O was added with stirring and the yellow solid was filtered using a Buchner fiannel. 



54 



The solid was collected and washed with another 100 mL aliquot of H2O. The solid 
was filtered again, and washed with CHCI3 2 times in a beaker and filtered. The solid 
was dried in a vacuum desiccator. The product's melting point corresponded well to 
the literature value of 210° C. lodometric titration, involving converting the PhIO 
product to Phi and I^ with HI and titrating with sodium thiosulfate, gave a purity of 
99%. 1 22 

PhIO (27.5 mg, 125 ^mol) was diluted in 25 mL CH^Cl^. The PhIO compound 
was only slightly soluble in CH^Cl^, so it was diluted and then sonicated for at least 30 
min. After sonicating, an amount of the cyclooctene was added and the mixture was 
stirred for about 1 min. Decane (97 ^L, 500 ^mol or 24 ^L, 125 ^imol), the internal 
standard, was also added with stirring. ImL samples of this mixture were used for 
both a blank run and a homogeneous run. In all homogeneous experiments, 1 ^iL of a 
1 mM solution of MnPO was added to the blank solution to investigate the epoxide 
yields with the porphyrin in solution. 




Screws to hold 
cells together 



Figure 2.3: Schematic of catalysis cell, side view. 



55 



Approximately 5 mL of the above oxidant/substrate solution was transferred 
into a small Erlenmeyer flask and from there loaded into the flow cells used for 
studying catalysis with the films. Also, blanks and homogeneous solutions were 
loaded into cells containing blank films (no catalyst) for studying product yields 
affected by the flow cell. 



Viton cell: 
Bottom plate 




Groove with 
Viton o-ring 



0.06" wide 
ledge 



Inlet/Outlet 
hole 



Viton cell: 
Top plate 




Figure 2.4: Schematic of catalysis cell, top view. 



The flow cells were built by the University of Florida machine shop. The cell 
was made from two blocks of Delrin into which was carved a cell with dimensions: 
0.96" X 1 .38" X 0.039" . Inlet and outlet tubes were place at the cell edges. One end 
of a 1' length of 1/16" ID Viton tubing (Cole-Parmer, Vernon Hills, IL) was 
connected to the inlet port and the other end was submerged in the reaction solution. 
A Cole-Parmer Masterflex peristaltic pump (model 7553-70) (6-600 rpm) with an 
easy-load head was used to introduce the solution into the cell, then the open end of 
the Viton tubing was removed from the solution and connected to the outlet port of the 
cell. The solution was circulated around the film using the peristaltic pump. 



56 



The reaction products were studied by GC. The GC instrument was a 
Shimadzu GC-17A (Columbia, MD) with a hydrogen flame ionization detector. A 1 
HL portion of the reaction solution was injected onto the 25 m, 0.025 mm ID RTX-5 
column (Crossbond, 5% diphenyl-95% dimethyl polysiloxane). The column was held 
at 50° C for 3 min and then ramped at 10° C min' for 15 min. 

Sensitivity factors (k) were determined using decane and o-dichlorobenzene as 
the internal standards. A series of runs were performed for both the cyclooctene 
(CyO) and the cyclooctene oxide (CyOO) standards. Equation 2.2 was used to 
calculate 'k' from the GC trace areas of the standard (AJ and the product (Acyoo) and 
the known sample weights (Wcyoo and wj. 



I^CyOO = (2.2) 



After an average k value was obtained for the CyO and CyOO, the catalysis yields 
were determined from the reaction mixture using Equation 2.3: 



VK-../V1 = 



_ ^Cyoo^sA:, 



yOO 



^CyOO = (2.3) 



Because the PhIO oxidant was rather insoluble, a series of 1 mL aliquots of a 
1.1 g L-' solution of PhIO in CH^Cl^ were dried and weighed. The final weights were 
1.1 mg ± 9%, assuring us that the amount of oxidant in each reaction was 
approximately the same. 

In order to compare nearly the same concentrations of catalyst in both the 
homogeneous and heterogeneous cases, the concentration of porphyrins in the films 



57 



was calculated to be, at most, 1 nmol. In was, however, difficult to keep this 
concentration constant. In the homogeneous reactions, 1 nmol of the corresponding 
non-amphiphilic MnPO was used. 

Imidazole is reported to improve the catalytic efficiency of the Mn-porphyrins 
in the presence of a peroxide oxidant, but for comparison, we also tried to see if 
imidazole would improve the catalytic efficiency in the presence of PhlO. For this 
experiment, 0.1 |a,mol of ImH was added to the blank and homogeneous solutions 
using a PhIO solution with 40 |amol CO, 5 i^mol PhIO and 20 )Limol decane. Also, 
films prepared by SA ImODPA and SA MnP4 were used in the flow cells with this 
same PhIO solution. The reactions were, again, stirred for 24 hr. 

2.3.2 Catalysis using peroxide oxidants 

In studying the catalysis of the epoxidation of cyclooctene (CyO) with H2O2, 
many different substrate to oxidant ratios were studied. The epoxide yields were 
greatest, however, when very dilute solutions of reactants were used to keep the 
proportion of reactants to catalyst near a factor of 10 and the substrate was used in 
excess. The starting materials, CyO and HjOj, were dissolved in 250 mL of CH2CI2 
with o-dichlorobenzene as the internal standard. ImL aliquots of this solution were 
used for the reaction blank and homogeneous reactions and contained 8 |amol CyO, 
0.2 )Limol H2O2, and 0.2 ^mol of o-dichlorobenzene. To the homogeneous reaction 
were added 0.4 i^mol ImH and 0.001 |amol of MnPO. The original solution was 
pumped into the flow cells using the peristaltic pump, and these reactions were 
allowed to stir for 24 hr at room temperature. 



CHAPTER 3 

PALLADIUM PORPHYRIN CONTAINING ZIRCONIUM PHOSPHONATE 

LANGMUIR-BLODGETT FILMS 



3.1. Background on Palladium Porphyrin Films 

Langmuir monolayers and LB films of the derivatized palladium tetraphenyl 
porphyrin molecules, palladium 5, 1 0, 1 5,20-tetrakis(2,3,5,6-tetrafluorophenyl-4- 
octadecyloxyphosphonic acid)porphyrin and palladium 5,10,15-tris(2,6- 
dichlorophenyl)-20-(2,3,5,6-tetrafluorophenyl-4-octadecyloxyphosphonic 
acid)porphyrin, referred to as PdP4 and PdPl, respectively, are described in this 
chapter. The central Pd metal is a four-coordinate, diamagnetic d* metal, and is co- 
planar with the porphyrin ligand. Being four-coordinate, there are no complicating 
axial ligands to consider. 

Porphyrins PdP4 and PdPl are substituted with four and one 
octadecylphosphonic acid groups, respectively. These molecules differ from many 
other porphyrin amphiphiles in that there is a hydrophilic group at the end of the alkyl 
chain substituent. In many literature reports, the porphyrin group is often the 
hydrophilic part ofLB film forming amphiphiles.9' Molecules PdP4 and PdPl 
(Figure 3.1) were designed to investigate whether porphyrins can be incorporated into 
metal phosphonate LB films. Also, because of their well-understood spectroscopic 
behavior, the PdP molecules were used to study how the orientation and aggregation 
of the porphyrin can be controlled in the deposited films. 



58 



59 



0(CH2)i8P03H2(R) 





0(CH2)i8P03H2 



Figure 3.1: Structures of A) PdP4 and B) PdPl . 



B 



Palladium polyhalogenated porphyrins were chosen based on the following 
considerations. First, palladium porphyrins are not demetallated in acidic conditions, 
and their diamagnetic character makes following the synthesis with NMR 
spectroscopy possible. Second, manganese and iron polyhalogenated porphyrins are 
well-known catalysts for the oxidation of hydrocarbons; '23 therefore, if palladium is 
replaced by one of these metals, the films can be used for catalytic applications. 
Third, pentafluorophenyl substituents on porphyrins allow straightforward 
functionalization of this ligand by aromatic nucleophilic displacement with an alcohol. 
In addition, the ether bonds are more stable toward hydrolysis and less hydrophilic 
than the ester or amide linkages usually used to tether alkyl chains on porphyrins. 
High hydrophobicity of the linkage is a requirement because competition with the 
polar phosphonic acid head-group should be avoided during the LB film preparation. 



■.^,^ 



60 



LB films of PdP4 and PdPl were formed incorporating a zirconium 
phosphonate network. The strong tendency of zirconium ions to crosslink the 
phosphonate groups precludes the normal deposition of organophosphonate 
monolayers with the metal in the subphase.28 Therefore, a previously developed 
three-step deposition procedure was used (Figure 2.2) as described in Chapter 2.28.38 
Both symmetric (PdP/Zr/PdP) and alternating (ODPA/Zr/PdP) films have been 
prepared in this way (Figure 3.2). 



a 



FT uTu S ii 




Hydrophobic Substrate 
= PdP1 and PdP4 I = OPA 



i 



Figure 3.2: Schematic of Pd-porphyrin films formed: a) alternating ODPA/Zr/PdP b) 
alternating ODPA/Zr/PdP:ODPA mixed film, c) symmetric PdP/Zr/PdP, d) symmetric 
PdP:ODPA/Zr/PdP:ODPA. 



Control over aggregation of the porphyrin chromophores is achieved through a 
combination of molecular design and the carefiil choice of the conditions for transfer 



61 



of the films. Aggregation is decreased or eliminated in the films of the tetra- 
substituted PdP4 when transferred at very high mean molecular area MMA and at high 
subphase pH. Mixtures of this amphiphile with ODPA transferred at high 
temperatures (40°C) and high MMA showed a similar decrease in inter-chromophore 
interaction. The four long-chain phosphonic acid substituents significantly aid the 
spreading of monomeric porphyrin species and the strength of the zirconium 
phosphonate interaction assures their isolation in the transferred films. In similar 
studies of the mono-substituted PdPl, aggregation was observed under all of the 
transfer conditions explored, indicating that none of the deposition procedures 
overcome the tendency of the molecules to aggregate. 

3.2. Results 

3.2.1. UV-vis of Palladium Porphyrin Solutions 

The palladium porphyrins show spectral responses in the UV-vis consistent 
with hypso-type metallo-porphyrins. For each porphyrin, a strong Soret Band (or B 
Band) is present above 400 nm and two Q Bands are centered around 550 nm.80.i24 
Solution studies of the porphyrins PdP4 and PdPl were performed in ethanol and 
chloroform, and the absorbance dependence on concentration was investigated. 
Solutions ranging from IQ-i i M to 10-6 M were studied (Figure 3.3). In CHCI3, the 
Soret Band was consistently at 410 to 41 1 nm for the porphyrin PdP4. In CHCI3, 
therefore, PdP4 shows no sign of solution aggregation. For porphyrin PdPl at 10"' 1 
M, the Soret Band absorbed at 41 1 nm; however, as the concentration was raised, the 
Band shifted to 414 nm. Because PdPl has only one long chain substituent, the 
likelihood of aggregation is increased; therefore, in CHCI3, the PdPl chromophores J- 
aggregate at high concentrations.76,92 Interestingly, the Soret Bands for both PdP4 



62 



and PdPl absorb at 41 1 nm at lO'i i M, so the long chains have no effect on the Soret 
Band of the non-aggregated chromophore. 



0.15 




0.00 



375 



400 425 

Wavelength (nm) 



450 



Figure 3.3: Solution UV-vis of Pd-porphyrins in CHCI3: A) PdP4 B) PdPl The 
absorbance scale refers to the lO"' M curve. The 10" M curve ha^ been enlarged for 
Band comparison. ^ 



The studies of the same molecules at identical concentrations in EtOH and 
water showed very different behavior (Figure 3.4). In EtOH at lO"' 1 M, both PdP4 and 
PdPl show Soret Bands at 414 - 415 nm. This peak is significantly to the red of the 
Soret Bands in CHCI3; however, it is known that more polar solvents tend to stabilize 



63 



the excited states in n-n* transitions, shifting this Band to lower energies.25.26 As the 
concentrations of both PdP4 and PdPl were raised, the Soret Band systematically 
shifted blue to 407 and 41 1 nm for the porphyrins PdP4 and PdPl, respectively, 
implying H-aggregation of the chromophores in EtOH.76.92 Going from EtOH to 
CHCI3 to H2O, at 10"* M, there is an obvious red shift in the X„^. This red shift does 
not correspond to a solvent polarity shift, but it does correspond to a shift in the 
solubility of PdP4. PdP4 is very soluble in EtOH and only slightly soluble in water. 



0.10 
0.08 


' 1 ' 


EtOH 




V 


CHCI3 

H,0 


<D 0.06 


/ 







'/ ^ \ 




c 


li/ 1 \ 




m 


1/ \ 




■e 0.04 


J ' \ 







jf * \ 




(/> 


// . \ 




J2 


/ / I 




< 0.02 


// \ \ 


- 


0.00 


-^ V 


^--^.-^^^^i^.^ 


._ 1 , 1 , 1 , r 



350 400 450 500 550 

Wavelength (nm) 



600 



Figure 3.4 Solution UV-vis of PdP4 in EtOH and water compared to CHCI3. 



3 11 Langmuir Monolayers of Palladium Porphyrins 

The room temperature pressure (D) vs. area (MMA) isotherm of PdP4 on water 
at pH 5.5 is shown in Figure 3.5. There is a measurable onset of surface pressure near 
220 A2 molecule- 1, followed by a gradual increase in pressure as the film is 
compressed, with an apparent phase transition giving a steeper rise in pressure near 



64 



1 15 A2 molecule-'. The MM A of the tetraphenyl porphyrin is 200 A2 molecule-' 
implying that at the onset, the tetrasubstituted porphyrin molecules are not aggregated 
or stacked.9' However, this arrangement is not stable to pressure, and as the film is 
compressed, the molecules are forced to rearrange. 



70 
60 
50 
40 
30 

8 ^ 

^ 10 

CO 







Pure PdP4 



50 100 150 200 250 

Mean molecular area (A^ molecule ^) 



Figure 3.5: Isotherms of PdP4, pure and mixed with ODPA (PdP4:0DPA), on a water 
subphase. 



The change in the aggregation of PdP4 during compression can be observed 
with reflectance UV-vis spectroscopy of the Langmuir monolayer (Figure 3.6). As the 
film is compressed from a MMA of 370 A?- molecule' through 220 A2 molecule', the 
>^max remains between 416 and 417 nm, similar to the X^ax observed for the non- 
aggregated porphyrin in EtOH. At areas between 220 and 100 A2 molecule-', the Soret 
Band shifts to 418 - 419 nm, and below 100 A2 molecule' the Soret Band shifts 
fiirther to near 421 nm. The shift in the Soret Band suggests a change in the 



65 



interaction of the chromophores at different pressures. At MMA larger than and 
comparable to the size of the chromophore itself, the porphyrin rings cannot be 
aggregating to any significant extent or the onset of surface pressure would occur at 
lower areas. The red shift of the Soret Band as the area is decreased indicates 
enhanced chromophore aggregation at lower MMA. 




380 400 420 

V\feveiength (nm) 



Figure 3.6: Reflectance UV-vis of PdP4 on water subphase. 



In contrast to PdP4, the U-A isotherm of PdPl (Figure 3.7) indicates that these 
molecules aggregate even in the absence of applied pressure. No significant increase 
in surface pressure is seen until areas below 120 A2 molecule-'. The pressure rises to 
only 5 mN m-' at 60 A2 molecule-', below which the pressure increases until the film 
collapses below 36 A2 molecule-'. The isotherm cannot reflect a true molecular 
monolayer, but rather results from the compression of aggregates at the water surface. 



66 



Evidence of aggregation at all MMA is seen in the reflectance UV-vis spectra. Figure 
3.8 shows the Soret Band as a function of MMA from greater than 120 A2 molecule-' 
to film collapse at 36 A2 molecule-i. The Soret Band does not shift during 
compression, and the X^ax of 426 nm indicates that the porphyrins are aggregated at 
each stage of the isotherm. 



E 
Z 
E, 

<u 

k. 

V) 
(/} 




0) 

o 

3 
03 




20 40 60 80 100 120 

Mean molecular area (A^ molecule"") 



Figure 3.7: Isotherms of PdPl, pure and mixed with ODPA (PdPl :ODPA) on a water 
subphase. ' 



A common procedure for enhancing the stability and processibility of unstable 
Langmuir monolayers, and to reduce aggregation, is to mix the amphiphile of interest 
with a good film-forming amphiphile.33,88-90 in this pursuit, both of the porphyrins 
were mixed with ODPA, which is a well-studied amphiphile that forms a liquid- 
condensed phase on the water surface and easily binds to an exposed Zr-phosphonate 
surface. As the percentage of ODPA is increased, the isotherms increasingly take 



on 



«7 



characteristics of the liquid-condensed phase of ODPA, although features present in 
the isotherms of the pure porphyrins are also present in the isotherms of the mixed 
films (Figure 3.5 and 3.7). 



0.12 


r 


1 


1 ■ 1 ' 1 — 




. 50 


1' 1 


0.10 


-^^40 


A/tU___ r . 




. t 30 


V 


// 1 )\ 


n, 008 


-| 20 


V 


//V\ft — ^ 


S 


.^10 


M_F U 


/Jv U^M-^d 


^ 0.06 


- 


— 1 — i_i . 


■,oy^(/^ 


SI 




25 50 75 100 125 1£ 


o 0.04 


MMA (A' molecule') 


£/^^^ ■ 


"^ 0.02 
0.00 


,, i,aiitf|!l^^ 


^^^^^%L! 


¥f$mm0i(iimi^^^^^'^^^ 


1 


-0.02 


1 1 1 1 1 


1 1 , 



360 380 400 420 

Wavelength (nm) 



440 



Figure 3.8: Reflectance UV-vis of PdPl on water subphase. 



In addition, the collapse pressure increases with the concentration of ODPA indicating 
that the films become more stable as ODPA is added. However, diluting the porphyrin 
film with ODPA does not appear to greatly affect the aggregation. Reflectance UV- 
vis of a Langmuir monolayer of a 1 :9 mixture of PdP4 with ODPA is shown in Figure 
3.9. The Xmax shifts from 415.5 nm at high MMA to 420 nm as the film is 
compressed, just as it does in the films of pure PdP4 (Figure 3.9). However, the 
porphyrins do not appear to be aggregated in the mixed film at high MMA. 



68 



0.03- 



0.02- 



0.01 



0.00 



-0.01 




V\^velength (nm) 



440 



Figure 3.9: Reflectance UV-vis of 10% PdP4: 90% ODPA on a water subphase. 

The molecular areas in Figures 3.5 and 3.7 are weighted averages of the 
porphyrin and ODPA molecules. The MMA of the porphyrin molecules in the mixed 
films can be calculated using Equation 3. 1 : 89 



-"MIX 



{N+\) 



(3.1) 



where Smix is the MMA of the mixture determined from the isotherm, SpoR is the 
MMA of the porphyrin within the mixed films, Sqdpa is the MMA of the ODPA 
amphiphile in pure ODPA films, and N is the molar ratio of ODPA to porphyrin. 
SpoR was calculated in the ODPA mixtures of each porphyrin at pressures of 5 mN 
and 15 mN m-' and the results are plotted in Figure 3.10. 



m- 



69 



300 
225H 



2 150- 



co 



■ 5 mN m"^ 
• 15mNm'^ 




2 4 6 8 10 
N (OPA/POR) 



75- 


B 


■ ^"~**** 


— ■ 1 ■ 1 ■ 

■ 5 mN m'^ 
• 15mNnT' 


50 


^_ 




^^^^^ 






• ^^""-"^ 




25 






^^^ 


0-J 




— 1 " 1 — 





2 4 6 8 10 

N (OPA/POR) 



Figure 3.10: Mean molecular area vs. ratio of ODPA/Porphyrin: A) PdP4, B) PdPl. 

If the ODPA diluent were breaking apart the preferred organization of the 
porphyrins in the films, SpoR would increase as the aggregates separate. The decrease 
in SpoR in the mixed films suggests that either the porphyrin chromophores are 
reorienting in the mixed films or aggregation actually increases in the mixed films. 
However, it does not appear that porphyrin aggregation decreases in the mixed 
monolayers. 



3.2.3 Langmuir-Blodgett Films 

LB films of PdP4 and PdPl were prepared using the deposition procedure 
described in Figure 2.2. The stepwise deposition allows fabrication of both symmetric 
films, where the template and capping monolayers are the same, and alternating films, 
where the two monolayers in the bilayer are different. Both types of films were 
prepared for each porphyrin (Figure 3.2). It has been shown that a zirconated ODPA 
template layer fi-equently provides the best substrate for transferring a capping layer.65 



70 



The extremely well organized and oxophilic surface allows deposition of almost any 
phosphonic acid monolayer, including those that are not stable monolayers and would 
normally not transfer. Monolayers of PdP4 and PdPl were transferred at a range of 
temperatures, pressures and subphase pHs (Tables 1 and 2). Films of the porphyrins 
mixed with ODPA were also transferred under a variety of conditions. Under some 
conditions, perfect, organized monolayers were obviously not formed, but the films 
could be transferred onto solid supports and studied. 

3.2.3.1 Films of compound PdP4 . To form alternating films of PdP4, the 
Langmuir monolayers were transferred as capping layers onto zirconated ODPA 
template layers. Films were transferred at different surface pressures and the Soret 
Band of the transferred films was used to monitor differences in chromophore 
aggregation in the deposited films. The UV-vis spectrum of a film transferred at 130 
A2 molecule-' (15 mN m-') is shown in Figure 3.1 1, where the A-max of the Soret Band 
appears at 420 nm, significantly red-shifted from any of the solution spectra of PdP4. 
The red-shift suggests increased aggregation, which is expected because at such a 
small MMA, the chromophores must be either tilting perpendicular to the surface and 
organizing side-by-side, or sliding over one another to form bilayers or multilayers. 
Polarized UV-vis spectroscopy indicates the porphyrins are oriented parallel to the 
surface, implying the latter arrangement. 

Layers of PdP4 were also transferred at 190 (12mN m-') and 
300 A2 molecule-' (Figure 3.1 1). The Soret Band shifts to 418 nm for the film 
transferred at 190 A2 molecule-' and to 416 nm for the film transferred at 300 A2 
molecule-', indicating less aggregation in films transferred at high MMA. At these 
larger MMA, the porphyrin chromophores should be lying flat at the air-water 
interface with little aggregation and they appear to remain non-interacting when 
transferred. 



71 



Table 3.1: UV-vis data from symmetric and alternating films of PdP4. X^^ is given 
for monolayers, and interlayer thickness is given for multilayers of films transferred 
under a variety of transfer conditions. 



Film 


Transfer 


nof 


pH*** 


Temp 


^max 


thickness 




Area 


Transfer 




CC) 


(nm) 


(A) 




(A' mol.-')* 


(mN/m) 










0PA/Zr/PdP4 


300 




5.5 


23-25 


416 




OPA/Zr/ PdP4 


190 


4 


5.5 


23-25 


418 


42 


OPA/Zr/ PdP4 


180 


5 


5.5 


23-25 


418 




OPA/Zr/ PdP4 


130 


15 


5.5 


23-25 


420 




OPA/Zr/ PdP4 


100 


25 


5.5 


23-25 


420 




OPA/Zr/ PdP4 


90 


35 


5.5 


23-25 


420 




OPA/Zr/10%PdP4 


37 


5 


5.5 


23-25 


415 




OPA/Zr/10%PdP4 


33 


15 


5.5 


23-25 


418 


47 


OPA/Zr/25% PdP4 


50 


15 


5.5 


23-25 


420 




OPA/Zr/50% PdP4 


74 


15 


5.5 


23-25 


420 




OPA/Zr/ PdP4 


300 




9.4 


23-25 


416 




OPA/Zr/ PdP4 


190 


4 


9.4 


23-25 


416 




OPA/Zr/ PdP4 


300 




11.1 


23-25 


414 




OPA/Zr/ PdP4 


85 


15 


5.5 


40 


419 




OPA/Zr/ PdP4 


160 


4 


5.5 


40 


417 




OPA/Zr/25% PdP4 


50 


15 


5.5 


40 


415 




OPA/Zr/25% PdP4 


60 


4 


5.5 


40 


415 




OPA/Zr/10%PdP4 


35 


15 


5.5 


40 


415 




OPA/Zr/10%PdP4 


50 


4 


5.5 


40 


415 




PdP4/Zr/ PdP4 


190 


4 


5.5 


23-25 


416 




PdP4/Zr/ PdP4 


130 


15 


5.5 


23-25 


418 




10%PdP4/Zr/10% 


37 


5 


5.5 


23-25 


418 


47 


PdP4 















* Area of the chromophore and diluent as determined fi-om Figure 3.5 (isotherms) 
** Corresponding pressure from Figure 3.5 (isotherm) 



** 



* pH of nano-pure water fi-om filtration system is about 5.5 



72 



8 







1 ■-' 1 1 1 . 1 .— 


0.04 


- 


A ISOA^mdecule' 

r,\ 300 A' molecule-' 

/ '1 


0,02 


/ 


/ M 

/ '1 

/ M 

/ '\ 

/ ' \ 


0.00 


^ 


1 . 1 . t 



350 



400 450 500 550 

Wavelength (nm) 



600 



Figure 3.11: Transmission UV-vis of PdP4 films transferred at high and low MMA. 
Absorbance scale corresponds to the film transferred at 300 A^ molecule"'. The 
absorbance for the film transferred at 130 A^ molecule' has been divided by 10. 

As the pH was raised, the amphiphiles became slightly more water-soluble and 
the monolayer was increasingly susceptible to creep. However, films of PdP4 
compressed to 300 A^ molecule-i were deposited onto zirconated ODPA templates 
fi-om subphases of pH 9.4 and 1 1 . 1 . As the pH increased, Pimax of the Soret Band of 
the transferred film decreased to 414 nm for the film deposited at pH 1 1.1. This was 
the lowest value of X-max, and therefore, the least aggregated LB transferred film of 
PdP4. The A,niax of this Soret Band corresponds to that of chromophore PdP4 in EtOH 
at 10-'2 M which is believed to be non-aggregated. 

Consistently, D = 1 ± 0.02 when measured at 0° incidence, indicating no 
preferred in-plane orientation of the chromophore in the PdP4 and PdPl films. 
However, in all films, D ^ 1 when measured at 45 ° incidence. For films transferred at 
high surface area, it is expected that the porphyrins should lie flat with all four 
phosphonates tethered to the surface. Indeed, this is observed for films transferred at 



73 



190 A2 molecule-i and 300 A2 molecule-i where the tilt angle, 0, with respect to the 
surface normal is observed to be 90°. Interestingly, the porphyrins also appear to lie 
parallel to the surface in the films transferred at 130 A2 molecule-' where is also 
measured as approximately 90°. This result implies that in films transferred at areas 
smaller than the MMA of the flat porphyrin macrocycle, the molecules overlap, 
stacking in bilayers or multilayers but with very little change in the tilt angle. There is 
a larger uncertainty, possibly ± 10°, in the measurement as the tiU angles near 90'';26 
however, these results confirm that the chromophores are lying approximately flat in 
all of the films in this study. 

Multilayers of the alternating 0DPA/Zr/PdP4 films can be deposited and X-ray 
diffraction confirms the layered nature of the films. Two or three orders of the (00/) 
Bragg peaks can be observed in each case. Films transferred at 190 A2 molecule"' 
have a bilayer thickness of 42 A, which is smaller than the 48 A thickness seen in pure 
ODPA/Zr/ODPA bilayers,28 suggesting that the 18-carbon tethers of PdP4 are not 
fully extended in the alternating films. For the film transferred at 130 A2 molecule"', 
the bilayer thickness increases to 47 A as the tetrasubstituted chromophores begin to 
overlap. 

Symmetric bilayers of PdP4/Zr/PdP4 fabricated according to Figure 2.2, were 
also studied. Porphyrin PdP4 could be transferred on the down stroke onto a 
hydrophobic substrate under a variety of conditions. After zirconation, deposition of a 
capping layer of PdP4 results in a symmetric bilayer. The Soret Band is very similar 
to that from alternating films deposited at the same area per molecule, and polarized 
UV-vis indicates the porphyrins are also lying parallel to the surface. However, the 
layers are poorly organized, as (00/) Bragg peaks could not be seen in diffraction fi-om 



74 



9-bilayer films. It is probably poor organization in the template layer of porphyrin 
PdP4 that is responsible for the lack of a well-defined layered structure.38 

Mixed monolayers of PdP4 with ODPA were transferred onto ODPA templates 
at different points along the surface pressure vs. area isotherms as shown in Table 1, 
and the aggregation of the porphyrin in the transferred film parallels that seen in the 
films of the pure porphyrins. For films transferred at pressures of 15 mN m-', the 
Soret Band appeared at 420 nm, shifting to 415 nm when transferred at pressures less 
than 5 mN m"' which, again, corresponds to the non-aggregated form seen in EtOH. 
In all cases, polarized UV-vis indicates that the porphyrins orient parallel to the 
surface. 

The mixed monolayers show an interesting effect with increased temperature. 
A mixed monolayer of 10% PdP4 with ODPA transferred at 15 mN m-' on a subphase 
heated to 40° C shows a Soret Band X^ax of 415 nm, shifted fi-om 420 nm for the same 
film deposited at room temperature. As the subphase is heated, the aggregates appear 
to break-up in the mixed film. A similar effect is not seen on the pure films of PdP4. 
It appears that the ODPA plays a role in breaking up the aggregated domains at higher 
temperatures. 

Films of PdP4 were also prepared by the SA technique. After the zirconated 
ODPA template had been exposed to a PdP4 solution in EtOH/HjO for 2 hr, the 
porphyrins were successftilly incorporated into these films. The Soret Band appeared 
at 414 nm, which then shifted to 41 1 nm after 60 min rinsing in hot CHCI3. The X„^ 
in the SA film was the closest of any of the PdP films to that seen in the dilute 
solution. Therefore, it appears that non-aggregated assemblies of PdP4 are easily 
obtained by self-assembly (Figure 3.12). However, the overall absorbance intensity of 
these non-aggregated films is lower than observed in the films transferred by the LB 
technique at high MMA. 



75 



0.015 



8 0.010 

c 



0.005 



0.000 



-0.005' ' L 



SAPdP4 
rinse 60 min 




-J 1 u 



350 375 400 425 450 475 

V\fevelength (nm) 



Figure 3.12: UV-vis of SA PdP4 films rinsed in hot CHCL. 



3.2.3.2. Films of compound PdP 1 . The ll-A isotherms and the reflectance 
UV-vis experiments described above indicate that the molecules of PdP 1 aggregate 
upon spreading, and this aggregation is preserved in the transferred films. In contrast 
to PdP4, the monophosphonate PdPl is only slightly influenced by attempts to break 
up the aggregates by changing the deposition conditions. The UV-vis spectrum of a 
capping layer of PdP 1 transferred at 52 A^ molecule-' (12 mN m-i) is shovm in Figure 
3.13, where the Soret Band appears at 426 nm, consistent with the value observed in 
the reflectance spectrum taken fi-om the water interface. The shape of the Soret Band 
does not change for films deposited at higher MMA, higher temperatures, or in 
mixtures with ODPA. The peak position shifts only slightly (Table 2). The 
orientation of the chromophores were also unaffected by the deposition conditions. 
Polarized spectra consistently give tilt angles of 90°, corresponding to the porphyrins 
lying flat. 



76 




400 500 

Wavelength (nm) 



600 



Figure 3.13: Transmission UV-vis of films of PdPl transferred at high and low 
MMA. 



X-ray diffraction from alternating films of PdPl transferred at 52 A^ 
molecule-' onto a zirconated ODPA template gives a layer thickness of 61 A (Table 
3.2). This thickness is larger than that of the alternating films of 0DPA/Zr/PdP4 or 
ODPA/Zr/ODPA bilayers.28 Since optical spectroscopy indicates the molecules lie 
flat, the enhanced thickness of the layer suggests they transfer as stacked bilayers or 
multilayers. Further evidence for this arrangement comes fi-om the film stability 
studies, described below, which indicate that part of the transferred film of porphyrin 
PdPl is physisorbed to the surface. 



77 



Table 3.2: UV-vis data from symmetric and alternating films of PdPl. X„^ is given 
for monolayers, and interlayer thickness is given for multilayers of films transferred 
under a variety of transfer conditions. 



Film 


Area of 


nof 


Temp 


^max (nm) 


thickness 




Transfer 


Transfer 


CC) 




(A) 




(AVmolecule)* 


(mN/m) 


23-25 


426 




OPA/Zr/PdPl 


73 


4 




OPA/Zr/PdPl 


52 


12 


23-25 


426 


61 


OPA/Zr/ PdPl 


41 


25 


23-25 


428 


.^_ 


OPA/Zr/ PdPl 


38 


30 


23-25 


428 


—. 


OPA/Zr/ PdPl 


36 


35 


23-25 


428 


-^ 


OPA/Zr/10% 


27 


12 


23-25 


426 


48 


PdPl 












OPA/Zr/25% 


30 


12 


23-25 


426 


__ 


PdPl 












OPA/Zr/50% 


38 


12 


23-25 


426 


•^» 


PdPl 












OPA/Zr/ PdPl 


300 




23-25 


416 




OPA/Zr/ PdPl 


190 


4 


23-25 


416 


.^^ 


OPA/Zr/ PdPl 


300 




23-25 


414 


__ 


OPA/Zr/ PdPl 


73 


4 


40 


424 




OPA/Zr/10% 


30 


12 


40 


424 




PdPl 












OPA/Zr/10% 


26 


20 


40 


425 




PdPl 












OPA/Zr/25% 


30 


12 


40 


424 




PdPl 












PdPl/Zr/PdPl 


52 


12 


23-25 


426 


^^ 


10% PdPl /Zr/ 


26 


20 


23-25 


426 


48 


10% PdPl 













* Area of the chromophore and diluent as determined from Figure 3.7 (isotherms) 

** Corresponding pressure from Figure 3.7 (isotherm) 

*** pH of nano-pure water from filtration system is about 5.5 



n 



3.2.3.3. Film stability . Zirconium phosphonate LB films are insoluble in 
organic solvents due to the cross-linking within the zirconium-phosphonate extended 
network.28 In order to monitor how well the porphyrin layers bind to the zirconated 
template, transferred films of PdP4 and PdPl were rinsed with chloroform in a Soxhlet 
extractor. Figure 3.19 shows the Soret Band absorbances as a function of washing 
time in hot CHCI3. Figure 3.14 shows that none of the film of PdP4 is washed away 
after 1 hr in chloroform, suggesting that all of the molecules are tethered to the 
zirconated ODPA template. The same result was obtained for films of PdP4 deposited 
at both higher and lower pressures or in mixtures with ODPA. 

In contrast, the absorbance of the LB films of PdPl exposed to CHCI3 
decreased significantly due to the desorption of chromophores. The absorbance 
leveled off after 20 min, to a value corresponding to the truly surface confined 
chromophores (Figure 3.14). This result suggests that the stacked layers of 
chromophores in the porphyrin PdPl films were partially physisorbed on the surface. 



0.125 

0.100 



8 



TO 0.075 
o 

(0 
< 

0.025 



0.000 



^— « — » D— Q -« — 

— ■^*— >< 



10 20 30 40 50 60 

Time in Soxhlet (min) 



Figure 3.14: Absorbance of Soret vs. time rinsed in hot CHCI3: a) PdP4, b) PdPl . 



79 



3.3. Conclusions 

The behavior of the monosubstituted and tetrasubstituted porphyrins is very 
different on the water surface, and the differences are carried over to the transferred 
films. Porphyrin PdPl spreads on the water surface to a limited extent. Our proposal 
for how the molecules behave on the water surface is shown schematically in Figure 
3.15. Optical spectroscopy indicates that PdPl aggregates, but the ti-A isotherm and 
X-ray diffraction from the transferred layers suggest that the aggregates are only a few 
molecules thick. The aggregates are present at both high and low MMA and can be 
transferred, as aggregates, onto the zirconated ODPA templates. Some molecules 
from each aggregate chemisorb to the zirconated surface through zirconium 
phosphonate linkages, but some are physisorbed as part of the preformed aggregates. 
When exposed to hot chloroform, the physisorbed part of the film is dissolved away. 



1 , 1 L^ -, I 




I — ~i r—p 



20 40 60 80 100 120 140 
MMA (A' molecule') 



Figure 3.15: Illustration of orientation and packing of PdPl films transferred at high 
and low MMA. 



so 



Chemical modification with four alkylphosphonic acid sidegroups allows the 
porphyrin to spread completely. Porphyrin PdP4 spreads to a monolayer thick film at 
high MMA, and as the film is compressed, an increase in surface pressure is registered 
near 200 A^ molecule-', corresponding to the area of the flat porphyrin macrocycle. 
However, the side-by-side arrangement of the porphyrin chromophores is not stable as 
the pressure is increased and the film rearranges, with the molecules sliding over one 
another to form multiple chromophore layers. This behavior is illustrated in Figure 
3.16. 




100 150 200 250 

MMA (A^ molecule^) 



Figure 3.16: Illustration of orientation and packing of PdP4 films transferred 
at high and low MMA. 



The films of PdP4 can be transferred at high MMA onto zirconated ODPA 
templates to form monolayer or submonolayer films of the porphyrin chemisorbed to 
the surface with the chromophore ring oriented parallel to the surface. Films of PdP4 
can also be transferred at lower MMA, where the reflectance UV-vis indicates the 



81 






porphyrins are interacting. Analysis of the transferred films suggests that the 
porphyrin chromophores are lying flat and overlapping each other to form layers that 
are a few molecules thick. In contrast to the films of PdPl, all of the molecules in the 
aggregated films of PdP4 appear to be chemisorbed to the surface. None of the film is 
lost during rinsing with hot chloroform. The different behavior probably results from 
the fact that four phosphonic acid groups increase the chance of each molecule 
bonding onto the zirconium phosphonate network. Any orientation of the porphyrin 
macrocycle will direct at least one alkylphosphonic acid side chain toward the surface. 
Also, all of the phosphonic acids have the potential to reach the water surface at high 
MMA. Forcing a strongly hydrophilic group off of the water surface requires more 
energy than reorganization of the alkyl chains or shifting the chromophore 
interactions. 

Whether the Langmuir monolayers are transferred intact or reorganized during 
film transfer is not yet completely clear. At high MMA, molecules of PdP4 lie flat on 
the water surface and this arrangement appears to be preserved in the transferred films 
based on the similar X^^. However, when the films are transferred at lower MMA, 
where the films are clearly aggregated, there could be some rearrangement. There is a 
significant driving force for forming zirconium phosphonate linkages, and the 
aggregates could rearrange during transfer to fiarther maximize interactions with the 
zirconated surface. While the porphyrins are clearly oriented parallel to the surface in 
the transferred film, providing each porphyrin the chance to form multiple zirconium 
phosphonate bonds, it is not known if the molecules aggregate the same way on the 
water surface. 

The LB procedure used in these studies takes advantage of the binding energy 
of the zirconium phosphonate continuous network and is shown to be quite versatile. 
Both symmetric and alternating layer films can be prepared. Use of the zirconated 



82 



ODPA template layer allows almost any phosphonic acid derivatized amphiphile to 
transfer in a capping layer.28.38,65,i25 Unlike conventional LB depositions, films of 
PdP4 and PdPl can be transferred onto the zirconated template layers at any surface 
pressure, allowing, in the case of PdP4, the arrangement of the molecules in the 
transferred films to be tuned by choice of the area-per-molecule at deposition. The 
films do not need to be stable Langmuir monolayers in order to transfer, as the driving 
force is formation of the zirconium phosphonate bonds. It is the strength of the 
zirconium phosphonate interaction, in particular the lattice energy associated with the 
zirconium phosphonate extended network, that is responsible for the exceptional 
stability of these non-traditional LB films.28.38 

Zirconium phosphonate LB films, like solid-state zirconium phosphonates, are 
insoluble in organic solvents and under most aqueous conditions. The inorganic 
network has also been shown to enhance the thermal stability of LB films. 1^6 The 
zirconium phosphonate inorganic extended network adds substantial stability to the 
films, which are insoluble under most organic and aqueous conditions. The methods 
developed here with the palladium tetraphenyl porphyrins can also be applied to other 
porphyrin systems, and in this way the vast array of physical and chemical 
characteristics of porphyrins, including catalytic activity, should be able to be 
incorporated in stable LB films. 






CHAPTER 4 
MANGANESE PORPHYRIN CONTAINING 
ZIRCONIUM PHOSPHONATE THIN FILMS 



4.1 Background 

Monolayer and film work using the molecule manganese 5,10,15,20- 
tetrakis(2,3,5,6-tetrafluorophenyl-4-octadecyloxyphosphonicacid)porphyrin, or 
MnP4, will be discussed in Chapter 4 (Figure 4. 1 A). For comparison, work done 
using a similar molecule without the four alkylphosphonic acid chains, manganese 
5,10,15,20-tetrakis(penta-fluorophenyl)porphyrin, or MnPO, will also be discussed 
(Figure 4. IB). The manganese porphyrins are structurally and chemically more 
complex than the palladium porphyrins. The Mn(III) central metal is a 5- or 6- 
coordinate d'* metal. 80, 127, 128 Depending on the ligand character, Mn(III) is either S = 
2 (high spin) or S = 1 (low spin). 129 Also, depending on the axial ligand or ligands, 
the Mn(III) may or may not be co-planar with the porphyrin ligand. Mn(III) also has 
an easily accessible lower oxidation state, which leads to significant metal/porphyrin 
electronic interactions.^o the Mn(lIl)-porphyrins have a tendency to form face-to-face 
dimers bridged through an axial ligand, 130 and the Mn(III)-porphyrins are vulnerable 
to demetallation under certain conditions. Therefore, film characterization using these 
molecules was much more complicated than with the Pd-porphyrins. 

To investigate the catalytic properties of manganese-porphyrin films, film 
preparation procedures involving the tethering of Mn-porphyrins to a metal 



83 



84 



phosphonate network were developed. This method involves the initial formation of a 
zirconated octadecylphosphonic acid (ODPA) template onto which a film of pure 
MnP4 can be SA or transferred via the LB technique (Figure 2.2). Including MnP4 in 
a zirconium phosphonate network provided films that were stable toward harsh 
organic conditions. Also, the strong oxophilicity of the zirconium for the phosphonate 
oxygens enabled the film preparation procedure to be easily altered and fine tuned and 
complete film characterization to be carried out. 



0(CH2)18P03H2(R) 





Figure 4. 1 : Structures of A) MnP4 and B) MnPO. 



The MnP4 molecule is similar to the PdP4 molecule described in Chapter 3, in 
which the manganese tetraphenylporphyrin (MnTPP) chromophore and the strongly 
hydrophilic phosphonate groups are separated by 18-carbon chains (Figure 4.1 A). This 
geometry allowed for the porphyrin to be sitting at the exterior of the film and 
available for catalysis while the phosphonates were buried in the hydrophilic region 
and available for binding to the stabilizing inorganic network. The incorporation of 
this network significantly improves the resistance of the film to typically destructive 



85 



forces such as solvent, heat, or time.'26 in addition, having four amphiphiUc chains on 
the porphyrin permits the formation of Langmuir monolayers of these materials 
without diluting the amphiphiles with a good film forming amphiphile such as stearic 

acid.33,88-90 

The MnP4 films were first investigated on the water surface in a Langmuir 
monolayer. An isotherm of this material showed significant film compressibility 
(Figure 4.2). Reflectance UV-vis showed that the porphyrins formed face-to-face 
dimers above ca. 10 mN m'', which were maintained upon transfer onto glass 
substrates. 

The procedure for MnP4 film formation was directed by the film 
characterization results. In most cases, evidence suggests that the phosphonic acid 
tethers on the porphyrins were able to bind to the metal phosphonate lattice. Film 
stability was monitored by UV-vis, which displayed no significant chromophore loss 
after 5 minutes in hot CHCI3 or CHjClj. Although the zirconium phosphonate lattice 
contributes no interesting physical phenomena to the final film, the strong oxophilicity 
of the phosphonate oxygens for the zirconium lattice allow for a wide variety of stable 
films to be formed. 

4.2 UV-vis Behavior of MnTPPs 

4.2.1. Solution Studies 

Both MnTPPs displayed electronic behavior in solution consistent with d-type 
hyperporphyrins. Mn(II)-porphyrins have absorption spectra similar to free-base 
porphyrins due to a lack of metal-porphyrin interaction; however, the absorption 
spectrum of the Mn(III)-porphyrin is quite different. According to Gouterman,^^ 
Mn(III)TPP is a classic d-type hyperporphyrin with extra absorption bands at higher 



86 



energies relative to A,n,ax- The MnP4 and MnPO with chloride axial ligands are 
spectrally consistent with d" porphyrins in high-spin configurations. '^i The strong 
absorption near 450-485 nm is often called Band V, due to the fact that this transition 
is not pure n-n* in nature but includes metal-porphyrin orbital mixing. However, 
traditionally, it is still often called the Soret Band.^o One strong band commonly seen 
to the blue of the Soret Band is called Band VI. A prominent peak, often observed 
specifically in the MnP4 UV-vis spectrum ca. 410 nm, is referred to as Band Va. 

Ligand effects and orientation or aggregation effects are reflected primarily in 
the shape or shift of the Soret Band and in the extinction coefficient (s).>32 Therefore, 
the behavior of this band was carefully monitored. Also, different solvents used for 
porphyrin investigations caused changes in the absorption spectra. If a coordinating 
solvent was used, the axial ligand was displaced by a solvent molecule causing a shift 
in the Soret Band and in the VA^I intensity ratio.80>'32 

4.2.1.1. UV-vis of MnPO in solution . A concentration study of MnPO in 
CH2CI2 showed that the >.max was consistently at 475 - 477 nm between 10-5 and 10-8 
M (s = 1.25 X 10' M"' m"'). These results suggest that the MnPO chromophore had 
little tendency to aggregate in these dilute solutions, and that the axial ligand was 
probably chloride. '32 In EtOH, the X^ax of the MnPO solufion was also constant over 
the same concentration range; however, the Soret Band was blue shifted to 454 - 456 
nm. According to Mu, the coordination of two axial methanol ligands to a MnTPP 
caused a 10 nm blue shift relative to the chloride bound moieties; therefore, it is 
believed that this blue shift is due to bis-EtOH binding. '33 Figure 4.2 shows the 
solution UV-vis of MnPO in EtOH and CHCl, at 10"' M. Not only is the Soret Band 
shifted, but the ratio of Band V to Band VI has also changed, indicating a change in 
the metallo-porphyrin ligand environment. 



«7 




400 500 

Wavelength (nm) 



600 



Figure 4.2: UV-visofMnPOinCHClj 



4.2. 1.2. UV-vis of MnP4 in solution . Similar solvent shifts were observed in 
the case of MnP4. In CH2CI2 or CHCI3 at 10'* M or lower concentrations, the Xmax 
appeared at 463 - 464 nm, whereas in EtOH or water, the symmetric Soret Band 
occurred at 454 nm (Figure 4.3). The peak at 454 nm corresponds to the MnPO Soret 
Band in EtOH indicating a similar axial environment, likely a bis-ethanol complex; 
however, a very clear band, Va, was present at 418 nm in the MnP4 solutions. 
Additionally, the ratios of Band V to VI were different for each different solvent, 
implying that coordinating solvents effect the porphyrin axial environment. 

The observed MnP4 spectral behavior in CHCI3 was very different from the 
spectral behavior in EtOH or HjO. At 10"* M, the X^^ occurs at 463 nm with a distinct 
shoulder present on the red side of the Soret Band. As the concentration of porphyrin 
in CHCI3 increased from lO-^ M to 10"^ M, the red shoulder became a distinct second 
peak at 477 nm (Figure 4.4). The two peaks are hereafter referred to as Vi and Vii for 



88 



the first and second Soret Bands. The peak at 477 nm was identified as the formation 
of a five-coordinate MnTPP CI structure in the case of MnPO in CHCI3, and therefore, 
it clearly represented the analogous structure in MnP4 (Figure 4.2). 



u. lu- 




n^ EtOH 


0.08- 


HjO^ 


fn 






1 \ .CHC13 


g 0.06- 


/ \ Va 7 


/ 1 \ 


C 




/ \ \ 


CO 




/ 1 \ 


■9 0.04- 




/ \ \ 







1 \ 


M 




1* \ 


X3 




I \ 


< 0.02- 


••■' 


V^k.==^><^ 


0.00- 


■ ' T ■ 1 • 



350 



400 450 500 550 

Wavelength (nm) 



600 



Figure 4.3 Solvent behavior of MnP4 in water, EtOH and CHCI3. 




350 375 400 425 450 475 500 525 

Wavelength (nm) 



Figure 4.4: UV-vis concentration study of MnP4 in CHCI3: a) 10* M, b) 10"' M. 



-yf — l.J*^"' 



89 



The obvious difference between MnPO and MnP4 is the presence of the four 
C,8-chain phosphonic acid tethers linked to the phenyl rings of the MnP4. Therefore, 
the peak at 464 nm is, as previously mentioned, probably due to intramolecular 
phosphonate binding. However, the structures of the two chromophores also differ in 
that MnPO has four pentafluoro-phenyl substituents, and MnP4 has an ether linkage to 
the alkyl chain at the para-position of the phenyl groups. For a more direct 
comparison, a compound incorporating similar /7-methoxy-tetrafluoro-phenyl 
substituents was prepared (MnP-4MeO). This compound's spectral behavior was 
identical to the MnPO compound, further indicating that phosphonate binding causes 
the Soret peak at 464 nm. 



OCH, 




MnP-4MeO 

4.2.1 .3. MnPO in solution with ethylphosphonic acid . Though the alkyl chains 
may have altered the chromophore interaction in solution, '34 the stronger UV-vis 
effects were likely due to the presence of the intramolecular R-P0(0H)2 ligands. The 
binding of the phosphonic acids to the central porphyrin metal was confirmed by 



90 



studying solutions of MnPO (1 x 10"' M in CHCI3) with ethylphosphonic acid at high 
concentrations. In pure MnPO, the X^^ was observed at 475 - 477 nm; however, when 
1 X 10"* M ethylphosphonic acid was added, a distinct blue shoulder became visible. 
At ethylphosphonic acid concentrations above 3 x 10" M (over 1000 times the 
porphyrin concentration) the dominant peak was the peak at 460 nm (Figure 4.5). 
Further, upon addition of phosphonic acid to MnPO solutions in CHCI3, an obvious 
peak emerged at 41 8 nm. Band Va was absent in the pure MnPO solutions, hence, this 
peak was probably related to phosphonic acid binding. 




400 450 

Wavelength (nm) 



500 



Figure 4.5: MnPO in CHCI3 (1 x 10"' M) with ethylphosphonic acid: a) pure MnPO, b) 
1x10- M ethylphosphonic acid, c) 2 x 10" M ethylphosphonic acid, d) 3 x 10"" M 
ethylphosphonic acid, e) pure MnP4. 



The peak at 41 8 nm could represent many different states of the porphyrin. It 
could be due to the oxidation of the central metal, to a spin state conversion, or even to 



91 



demetallation of the manganese porphyrin. This peak was observed in the absence of 
any strong oxidants indicating that this peak is probably not representing oxidation 
products. .\ , . • 

The electronic spectra of Mn(III)TPP CI in DMSO was presented by Hansen 
and Goff.129 The high spin moiety that was formed by the coordination of the DMSO 
solvent molecules showed a sharp Soret Band at 465 nm with two prominent higher 
energy bands at 375 and 397 nm. However, when the molecule was converted to a 
low-spin Mn(III)-porphyrin complex by the bis-axial ligation of imidazolate anions, 
the Soret Band shifted to 451 nm and broadened. Also, the band at 397 nm increased 
in intensity. '29 In a similar way, the phosphonic acid or phosphonate may have been 
behaving like the imidazolate anion and converting the MnPO from a high spin to a 
low spin complex. The spin-state of MnPO in solution was studied using the Evan's 
method (described in Chapter 2). The NMR results indicated that the MnPO with and 
without phosphonic acid ligands were in the same spin-state, therefore, the observed 
spectral behavior was due only to the ligand environment of the porphyrin. 

The peak at 418 nm aligned with the Soret Band of the corresponding free-base 
porphyrin in CHCI3 (g = 9.0 x 10^). The strong attraction between manganese and 
phosphonates is well known; therefore, irreversible binding of the phosphonate to the 
manganese may result in the demetallation of the porphyrin. The demetallation could 
be facilitated by the protic nature of the phosphonic acid. This band was often 
observed, but its relative intensity was not consistent. 

When the MnPO/ethylphosphonic acid solution spectra were overlayed with a 
solution spectrum of pure MnP4, the red shoulder, Band Vii of the MnP4 spectra, 
aligned with the original MnPO peak. Also, the blue shoulder. Band Vi, which 
emerged with the addition of the ethylphosphonic acid to a MnPO solution, was similar 
in energy to the original MnP4 peak. 



92 



That such high concentrations of ethylphosphonic acid were necessary to 
induce a change in the MnPO spectrum indicates that the five-coordinate MnTPP(Cl) 
structure was generally favored over the six-coordinate MnTPP(Cl)(PA) structure (PA 
= phosphonic acid). However, with the four phosphonic acids linked to the MnP4 
chromophore through the alkyl chains, the effective concentration of phosphonic acid 
in the vicinity of the metal was very high. Therefore, MnP4 in CHCI3 solutions favor 
phosphonic acid axial ligation. Considering the Soret Band in the above MnP4 
concentration study (Figure 4.4), it is apparent that this preference is stronger in the 
more dilute solutions of MnP4. 

4.2.1.4. MnP4 in solution with chloride ions . If the UV-vispeakat 460 nm 
demonstrated the porphyrin's propensity to bind phosphonic acids, it was important to 
determine if the phosphonic acid could be replaced with another ligand such as 
imidazole (ImH), or chloride. According to Arasasingham, the displacement of a 
water or hydroxy ion by imidazole is encouraged by the electron-donating nature of 
the nitrogen base ligand. "4 Unfortunately, no literature precedence was found on the 
strength of the phosphonate binding. Though chloride binding was implied in 
concentration studies of MnP4 in CHCI3, it was further tested by adding tetra- 
butylammonium chloride, /-BuNHj^ CI, to solutions of MnP4 at constant 
concentrations of 1 0^' and 1 0"'' M (Figure 4.6). Similar behavior was observed with 
added bromide ions, with Band Vii shifted to slightly lower energies. As the chloride 
concentration increased, the red shoulder at 477 nm seen in the original MnP4 spectra 
in CHCI3 became dominant and the blue shoulder, indicating phosphonic acid binding, 
disappeared completely. . ? . s . 



93 



1.4- 


— ' 1 ' 1 • 1 ■ 

t 


1.2- 


A 


1.0- 


M 


S 0.8- 


/^r\ 


1 0.6- 

CO 

^ 0.4- 


n y 


0.2- 


^^^^^ V 




^^=^^ \/ ^^^ 


0.0- 


1 ' ..-,.- , 1 , 



425 450 475 500 

Wavelength (nm) 



525 



Figure 4.6: Solution UV-vis investigation of MnP4's sensitivity to displacement of R- 
PO(OH)' by chloride at 1 x 10' M (CHCI3). The arrows indicate the changes in the 
intensity of the peaks as the chloride concentration changes from 0.0 M to 0.1 M while 
the concentration of MnP4 was constant. 



4.2.2. Langmuir Monolayers 

The monolayer behavior of MnP4 was studied using surface pressure (IT) vs. 
Area (MMA) isotherms on water (Figure 4.7). Investigating the tilt angle of the 
chromophores transferred at various surface pressures by polarized UV-vis gave some 
indication of the chromophore orientation in the monolayers. The fl-A isotherm 
showed a distinct n onset at ca. 250 A^ molecule"'. However, the approximate area of 
the chromophore itself is ca. 200 A2 molecule'. This large onset area implied that the 
alkyl chains were initially buckled so both the chromophores and the phosphonic acid 
groups were sitting on the water surface. The hydrophilic nature of the porphyrin, 
especially if the two axial positions are coordinated with water, makes this a viable 



94 



scenario. In the first region of the isotherm ('a' of Figure 4.7), the chromophores and 
the phosphonic acids remained on the water surface and were simply compressed. 




50 100 150 200 250 300 

MMA (A^/molecule) 



Figure 4.7: Isotherm of MnP4 on water subphase. 



During the plateau region of the isotherm ('b' of Figure 4.7), the chromophores were 
pushed off the water surface and the porphyrin rings started overlapping. There were 
likely some dimers forming and not all phosphonic acid end groups reached the 
surface. In region 'c' of Figure 4.7, the MMA was approximately 100 A^ molecule"', 
implying the presence of dimers, which were then further compressed until the 
"monolayer" collapsed. At this point, the film was likely not a true monolayer as the 
chromophores essentially formed a bilayer. 

The separation of the chromophores from the water surface was accompanied 
by overlapping, which was also indicated by a consequent shift in the Xmax detected by 
reflectance UV-vis (Figure 4.8). In region 'a', the X^^ax was at 457 nm. While in 



95 



region 'b', the Soret Band shifted to and remained at 453 nm. The blue shift 
corroborates the ft)miation of face-to-face dimers. The mean molecular area of this 
spectral shift indicated that dimers began forming even before all of the chromophores 
were pushed off of the water surface. 



o 

c 

O 
(A 

< 







1 ■■ 




I ■-■ — 


' 


U.U3 


30 
20 


V 




/Ajt"''™'' ■ 


0.02 


-^10 





^^ 


Jdj JfJMi Y ^nnA^mr^i ' 


0.01 


- 


100 


200 300 

MMA 


jJKiiJuVn 


ym ' 


0.00 


yjllnljl 


tdrail^ 


m)it^ 


0.01 


1 


wFr^ 


^ 



420 440 460 

Wavelength (nm) 



480 



Figure 4.8: Reflectance UV-vis of MnP4 on water subphase. 



4.2.3. Langmuir-Blodgett Films of pure MnP4 

4.2.3.1. Deposition of MnP4 from a pure water subphase. Using the LB 
technique, the MnP4 monolayers were transferred at various points along the isotherm. 
The transmittance UV-vis spectra of these films showed that the blue shift in the Xmax 
followed the shift observed in the reflectance UV-vis experiments on the water surface 
(Figure 4.9). Films transferred around 5 mN m"' had a X-max at 462 nm, while films 
transferred at pressures higher than 10 mN m"' were blue shifted to 456 nm. The Soret 
Band blue shift suggested that the chromophores were interacting as H-type 



96 



aggregates. The chromophore interactions observed on the water surface were similar 
to those observed in the deposited films; however, the X^^ at ca. 460 nm also implied 
that the chromophores were intramolecularly ligating phosphonic acids. 





' 1 ' 1 


1 1 1 1 

^30mN/m 


0.03- 


• L 


1 15mN/m :a 


§ 0.02- 

o 

^ 0.01- 


' c 


Ll2mN/m 
AL_10mN/m 


"^^^^^X^->/// 


\vJL^5mN/m 


. 


^^^^^//f 


^N|LlmN/m 


0.00- 


1 1 I ■— , 


, . 1 1 



350 400 450 500 

Wavelength (nm) 



550 



600 



Figure 4.9: UV-vis of MnP4 capping layers transferred onto ODPA/Zr at different 
surface pressures. 



The polarized UV-vis spectra were studied to better understand the orientation 
of the chromophores within the monolayer. In the case of a high- spin, 6-coordinate 
Mn(III)TPP, the metal typically lies within the porphyrin macromolecular plane. '27 
Therefore, the tilt angle of the chromophore with respect to the surface can be 
determined from polarized UV-vis as described in Chapter 2.25,26,135 When the 
monolayer was transferred in region 'a' of Figure 4.7, the tilt angle was determined to 
be ca. 90°. Within the plateau region of the EI- A isotherm, the tilt angle immediately 
after transfer was as low as 60°, but within minutes, the chromophore relaxed back to a 



97 



90° orientation relative to the surface normal. When transferred after the plateau 
region, at areas much smaller than the chromophore itself, the tilt angles were 
consistently ca. 90°, further confirming the presence of stacked rather than tilted 
chromophores. 

The stability of films transferred by the LB technique before, during, and after 
the plateau region of the isotherm were tested by exposing the films to hot CHCI3 for 
up to 60 min each. When transferred at 15 mN m'' (Figure 4.10A), the original Soret 
Band was at 455 nm. After rinsing, the Band Vi shifted to 463 nm and a second Soret 
Band appeared at 477 nm. This red shift in the band associated with phosphonate 
binding is probably due to elimination of H-aggregated, physisorbed chromophores. 
Band Vi observed before and after rinsing in the films transferred at high MMA, 
remained at the same energy indicating that there was no significant change in the 
chromophore interaction at this transfer area. Because chromophore interaction was 
expected to be low at this MMA before rinsing, it was not surprising that no shift was 
observed. In each case, the absorbance intensity decreased significantly during the 
first five minutes of solvent exposure, and then leveled off (Figure 4.10). The 
formation of two Soret Bands, Vi and Vii, makes it difficult, however, to truly assign 
the change in absorbance intensities to a removal of chromophores. 

The phosphonic acids in the MnP4 would probably have a stronger tendency to 
be on the water surface than bound to the porphyrin. Considering the strong tendency 
for the Mn-porphyrins to bind water in the axial positions, the chromophore could be 
very hydrophilic, promoting their tendency to lay on the water surface. This behavior 
was observed in MnP4 films transferred at high MMA. In films transferred at lower 
MMA, chromophore aggregation may have kept many phosphonates from binding to 
the zirconium network, leaving them available for binding the manganese. Therefore, 



98 



the phosphonate ligation was not as reversible when the films were transferred at 
higher pressures. 



^ 



0.03 



0.02 



0,01 



0,00 




I ■ I 




Initial 
1/ 

After 5 - 20 min 
in CHCI3 soxhiet 



350 400 450 500 550 

\Afevelength (nm) 



600 



Figure 4.10: LB films of MnP4 transferred at A) 15 mN/m and B) 5 mN/m rinsed in 



CHCI3. 



When LB films of MnP4 transferred at 300 A^ molecule"' (0.7 mN m"') were 
rinsed only in hot CH3CN, the band at 477 nm (Band Vii) was again observed (Figure 
4.1 1). The Soret Band at 464 nm corresponding to phosphonate binding conversely 
disappeared. These results suggest that hot solvents were able to eliminate the 
phosphonic acid binding leading to the stable five-coordinate MnTPP(Cl) structure, 
and that the appearance of this structure is not a result of chromophore aggregation. 



99 



0.004- 


1 1 ■ 1 


1 , -T-- 


g 0.002- 
§ 0.000- 

< 


before rinsina / 

/ 1 
/ ' 


\ 5minCH3CN ' 


/ ' 
/ ' 

fUTj f 1 

A/ ' » / ' 


\ .J\.r\j<f/\y^^ 


-0.002- 


f 1 


" 



350 400 450 500 

Wavelength (nm) 



550 



4.1 1 : MnP4 transferred by LB at 0.7 mN m"' and rinsed in CH3CN: A) transferred 
from a 0.5 mg mL"' solution. 



4.2.3.2. Transfer of MnP4 from a chloride ion-containing subphase . In order 
to avoid intramolecular phosphonic acid-manganese binding upon LB transfer, 
chloride ions were incorporated into the subphase at a 0.1 M concentration. The 
porphyrin was spread and compressed to 4 mN m"' for transfer onto a zirconated 
ODPA template. In the spectrum shown in Figure 4.12, Soret Bands were observed at 
461 nm and 476 nm. Clearly, there exist domains of five-coordinate MnTPP(Cl) and 
six-coordinate MnTPP(Cl)(PA) structures. 



100 



0.008 




350 400 450 500 550 

Wavelength (nm) 



600 



Figure 4.12: MnP4 transferred from 0.1 M [CI"] aqueous subphase at 4 mN m''. 



4.2.4. Self-Assembled films of MnP4 

4.2.4.1. Self-assembly from pure solvent . MnP4 films self-assembled from 
EtOH/HjO (9/1 mixture) or CH2CI2 showed a Soret Band at lower energies than 
observed in MnP4 LB films. The X^^^ was now at ca. 463 - 464 nm, which 
corresponds to the peak in MnP4 solutions and in LB films after rinsing off 
physisorbed and aggregated chromophores. Therefore, this peak has been attributed to 
phosphonic acid binding to non-aggregated metallo-porphyrins. When self-assembled 
films of MnP4 were rinsed in hot CHCI3, the red Soret Band associated with the 
MnTPP(Cl) increased significantly, and the peak associated with the six-coordinate 
MnTPP(Cl)(PA) decreased, again proving that hot solvents can remove the 
phosphonic acid ligands leaving the chloride ligand intact. : 



101 



However, this five-coordinate structure was not rigid. When the films were 
left to structurally relax overnight, the peak at 477 nm decreased in intensity and the 
blue shoulder became more intense. This reversible behavior indicates that the ligand 
environment, which is very sensitive to solvent and heat, is flexible (Figure 4.13). 




425 450 475 500 525 

• - Wavelength (nm) 



550 



Figure 4.13: MnP4 self-assembled from EtOH/H^O and rinsed in CHCI3. The legend 
indicates the spectra after rinsing, after being left overnight and the rinsed again over a 
three day period. 



When these films were rinsed in CH3CN, behavior similar to CHCI3 rinsing 
was observed. The band at 477 nm, again, increased and the band at 464 nm 
decreased in intensity with time in the hot solvent. Again, if the film was left 
overnight, the band intensities reversed (Figure 4.14). Hot CH3CN, therefore, also 
eliminated the phosphonic acid ligand and caused the formation of the five-coordinate 
MnTPP(Cl). When the film relaxed, the phosphonic acids had a tendency to bind 
again to the central metal forming the six-coordinate MnTPP(Cl)(PA) structure. 



102 




350 400 450 500 550 600 

Wavelength (nm) 



Figure 4.14: SA MnP4 films with rinsing in hot CHjCN. 



All of the described Band Vii intensity increases were observed after rinsing in 
hot solvents. If, as was expected, the axial environment of the porphyrin was 
associated with the mobility of these anions in solubilized films, the effect should be 
even more pronounced while these films were in solution. To study this, MnP4 films 
were self-assembled onto substrates that were cut to fit inside a cuvette filled with hot 
solvent, and the UV-vis was taken immediately after the solvent was heated. The 
results were, as predicted, exaggerations of what was observed when the film was 
removed from the solvent. Figure 4.15 shows that after 5 min in hot solution, the ratio 
VA^I was ca. 1 .0, and dropped drastically up to 60 min in hot solvent as Band VI 



103 



increased in intensity and area. But, immediately after the film was removed from the 
solvent, Band VI dropped leaving the VA^I ratio, again, greater than 1 .0. 



0.09 




350 



400 450 

Wavelength (nm) 



500 



Figure 4.15: UV-vis response of a SA MnP4 film during rinsing with hot CH3CN. 



Polarized UV-vis results were obtained from the self-assembled films of the 
MnP4. The chromophore tilt angle was 90° at all self-assembly times. The self- 
assembly of the MnP4 was facilitated by the strong binding between the zirconated 
phosphonate template and the phosphonic acids on the MnP4 molecules, and the 
position of the phosphonic acids on the periphery of the chromophore lends it to lie 
flat in the films. 

When the MnP4 was self-assembled onto a zirconated ODPA template, and 
these films were then rinsed in hot CHCI3 or CH3CN, the Soret Band conversion from 



..^ . . 



104 



460 nm to 477 nm was consistently observed. However, when these films were rinsed 
in hot EtOH, an anomaly occurred. Instead of the increase in the band at 477 nm, as 
was so commonly observed in CHCI3 and CH3CN, now a peak to the high-energy side 
of the original band at 454 nm grew in intensity. This peak aligns with the band seen 
for the MnP4 and MnPO in EtOH solutions, indicating that in the films, domains of the 
chromophores bind EtOH and therefore experience mobility in the axial positions 
(Figure 4.16). 









1 ' 1 ' 


1 ■,■■■■ 




0.020- 




Rinsing /"^ 
15min '"s ,' ' 


- 




0.015- 




\'' ~-'/^ 


Before 


(D 






' / \ 


^^ rinsing 









' / ' ' 


, 


c 

(0 


0.010- 




' / ' 


\ 


^ 






/ ^ 


» \ 


k_ 








* \ 









/ / 


\ \ 


V) 
J2 


0.005- 




' ' / 




<f 




^ 


y _^/^ 


\ >^ 




0.000- 










-0.005- 




' 1 ' 1 ■ 


1 I - 



400 



425 



450 



475 



500 



Wavelength (nm) 



Figure 4.16: UV-vis of MnP4 self-assembled films before and after rinsing in hot 
EtOH. 



4.2.4.2. MnP4 Self-assembly from chloride solutions. To avoid phosphonic 
acid ligating the porphyrin central metal and to promote the phosphonic acid binding 
to the zirconium network, the MnP4 was self-assembled out of a 1.4 x 10"^ M solution 
of porphyrin that was 0.1 M in ^BuNH3^ CI". Following a successfiil self-assembly, 



105 



the resulting film will appear completely hydrophobic. However, after the SA fi-om a 
chloride containing solution, the film did not appear totally hydrophobic indicating a 
potentially incomplete self-assembly. However, the ]V[nP4 was clearly present in the 
UV-vis. After rinsing in hot CHCI3, the Soret Band at 475 nm was present with a 
distinct peak at 460 nm. When these films were studied after a night in the desiccator, 
the Soret showed little change in shape. This behavior is consistent with a preference 
for the five-coordinate MnTPP(Cl) after rinsing (Figure 4. 1 7). Interestingly, after SA 
from a chloride containing solution, no peak was observed at 418 nm representing the 
absense of free-base porphyrin. The excess chloride presumably competes with the 
ligation of phosphonic acids and prevents their irreversible binding, which 
consequently prevents the demetallation of the MnTPP. 





■ 1 ' 1 

After r 


1 ' 1 


—I 


0.03- 


sax/ 


1 * -t . ■-'' 


-• -' *' 




/ 


1 ,\ After rinsing 




(D 


/ 


.1' ', . hot CH^CI^ 




0.02- 


/ / 


. 


c 


/ 1 


1 t 




(D 


1 1 






■e 


j 1 


\ 1 
\ 1 


• 





/ \ / ' 


\ 1 




^ 0.01- 


i/^^ % / ;' 


\ 1 


- 


< 


\;..,//' 


\ 1 




0.00- 




^—^ 


^ 



350 400 450 500 550 

Wavelength (nm) 



600 



Figure 4.17: MnP4 self-assembled from a 0.1 M chloride solution. 



106 



As an additional confirmation of the successful tethering of MnP4 to the 
zirconated ODPA template, XPS was performed. A clear peak was observed for F,s, 
Ni^and Mn2p3 electrons. The Zrj^j and Zrj^j and P2p3 were also observed. From the 
XPS it is apparent that the MnP4 has been incorporated into the film and that the fihn 
is stable to UHV. ■.;•'."! 




Binding Energy (eV) 



Figure 4.18: XPS of MnP4 SA film. The insert is an enlarged view of the same 
spectrum between 200 and 80 eV. 



4.4 Conclusions 



MnP4 molecules can be successfully incorporated into ultrathin films using both 
the LB and SA techniques. From the UV-vis perspective, these films appear very 
similar. In each film, there is a characteristic Soret Band. Typically, this band occurs 



107 



at 460 - 464 nm, which has been attributed to formation of a six-coordinate 
MnTPP(Cl)(PA) structure. Also, a band at 376 nm was assigned to Band VI, and 
Band Va at 418 nm was identified as corresponding to the presence of free-base 
porphyrins in the films. Further, rinsing in a solvent that can eliminate the 
intramolecular binding of phosphonic acid causes the Soret Band to shift to ca. 477 rmi 
representing the five-coordinate MnTPP(Cl). This shift is most pronounced in 
solutions in which chloride ions have been added indicating that excess chloride 
prevents the binding of the phosphonic acid. 

The band representing phosphonic acid binding demonstrated some reversibility. 
After films of MnP4 were rinsed in CHCI3 or CH3CN, the structure was that of 
MnTPP(Cl). When these same films were left to structurally relax, the UV-vis 
showed the reappearance of the band representing the MnTPP(Cl)(PA), accompanied 
by a decrease in the intensity of the MnTPP(Cl) peak. When the chromophores were 
adhered to the zirconated ODPA network by less than four of the phosphonic acid 
tethers, the non-bound phosphonic acid groups remained in central metal's vicinity. 
Therefore, displacement of the phosphonic acid ligands with hot solvents does not 
prevent them from re-ligating as the film conditions change. 

The SA MnP4 films have Soret Band absorbance intensities consistently between 
•'■ 0.015 to 0.02 absorbance units. This absorbance intensity corresponds to the LB films 
transferred at MMA just before the plateau region. When the porphyrin surface 
coverage is incomplete, as in the LB films transferred below 5 mN m'' and in the films 
self-assembled for very short times, the absorbance intensity was consistently around 
or below 0.01 absorbance units. Therefore, depending on the deposition conditions, 
the amount of chromophore incorporated into the films is reasonably consistent. 
Consistency in the chromophore loading is helpful in employing these films in 
catalysis studies. 



101 



Unfortunately, though the Soret Band at 477 nm is usually associated with 
chloride binding, Soret Bands around 460 nm can be associated with a number of axial 
ligands and chromophore interactions making difficult absolute characterization of the 
porphyrin's ligand and aggregation environment. However, ultimately, the ligand 
filling the axial position on the chromophore is less important than its lability. If the 
metal center of the porphyrin is available for oxidation, the catalyst will be active. 
Also, as will be discussed in Chapter 5, if the imidazole can displace the axially bound 
ligand, it can positively influence the catalysis. ■ * 



CHAPTER 5 

INCORPORATION OF AN IMIDAZOLE LIGAND INTO MANGANESE 

PORPHYRIN CONTAINING ZIRCONIUM PHOSPHONATE THIN FILMS 



5.1 Background 

As in Chapter 4, the chromophore catalyst studied was a tetraphenyl porphyrin 
para-substituted with four octadecylphosphonic acid groups (MnP4), which could 
tether the porphyrin directly to the zirconium surface. For comparison, the model 
porphyrin MnPO with no phosphonic acid chains was also examined. The heterocyclic 
ligand used for these experiments was an alkylphosphonic acid imidazole (ImODPA), 
which could also be easily attached to the zirconium surface. From the propensity of 
the imidazole to protonate in the presence of HBr during the ImODPA synthesis, the 
imidazole unit was the bromide salt upon film preparation (Figure 5.1). Deprotonation 
of the imidazole was attempted in order to facilitate its binding to the metallo- 
porphyrin. To study the porphyrin's UV-vis sensitivity to an imidazole ligand in 
solution, an imidazole with no alkyl chains (ImH) was also used for solution studies. 

The anticipated structures of the mixed MnP4/ImODPA and MnPO/ImODPA 
films are shown in simplified form in Figure 5.2. The chromophore is at the exterior 
of the film and available to catalyze the reaction of interest. The bulky chromophore 
leaves vacant sites available for the ImODPA, which is tethered to the zirconium- 
phosphonate network under and around the chromophore (Figure 5.2). Multiple 



109 



110 



J-^'. 



ImODPA are anticipated to be located below the chromophore plane; however, only 
one has the opportunity to bind at a time to the metallo-porphyrin core. 



A 0(CH2)i8PO(OH)2 (R) 

F^ A. /F 




Br" 



/ 
-N 



~N 



(CH2)i8 

P0(0H)2 




-N 



H 



Figure 5.1: Structures of A) MnP4, B) MnPO, C) ImODPA and D) ImH. 



Ill 





r 




MnP4 



ImODPA 



Figure 5.2: Schematic of MnP4 and ImODPA films. 



A variety of methods were employed in order to accommodate the porphyrin's 
large MMA and encourage its binding to the imidazole in the films. These formation 
procedures, which were motivated by a number of intentions, will be described in this 
chapter. First, the Mn-porphyrin/imidazole binding is in an equilibrium state. In order 
to encourage the imidazole binding, or at least its availability for binding, the 
imidazole had to be present in excess relative to the porphyrin. To insure excess 
imidazole, this layer was often prepared first by either the SA or LB method. The 
hydrophilicity of both the phosphonic acid and the imidazole chain terminating groups 



112 



of the ImODPA made this molecule impossible to transfer as pure LB films. 
However, the ImODPA could be mixed with a good film-forming molecule such as 
ODPA or HDPA (hexadecylphosphonic acid), and these LB films were transferred 
successfully. 

The film preparation methods fall into two general categories. First, the two 
components, ImODPA and MnP4, were assembled by either LB or SA in two separate 
steps. Examples include mixed ImODPA/HDPA LB films onto which MnP4 was self- 
assembled, ImODPA SA films followed by MnP4 SA, and MnP4 LB or SA films with 
ImODPA SA in the second step. The mechanism of the second SA step is less 
straightforward than the first. The phosphonic acid tethers on the second amphiphile 
must either find vacant or defect sites in the zirconium network, or actually replace 
already existing Zr-phosphonate bonds. Because this step is not a true self-assembly, 
it will also be referred to as the "substitution" step. The second category of film 
preparations involves the MnP4 and ImODPA being assembled in one step, either as a 
mixed SA or LB transfer. All of the described methods were successful in 
incorporating both components to some degree. The most facile involved SA of 
ImODPA followed by substitution of MnP4. 

Following the formation of the ImODPA film, MnP4 could be substituted from 
CHCI3, EtOH, or EtOH/water. Immediately after substitution, some MnP4s were 
physisorbed to the surface and were easily removed by rinsing in a hot solvent. The 
chromophores that remained contained at least one phosphonic acid tether bound to 
metal phosphonate base. Film stability was followed by UV-vis, and after the first 5 
min of exposure to hot solvents, no more significant chromophore loss was observed. 

One goal of this project was to understand the propensity of the imidazole 
ligand to bind to the porphyrin and to determine certain characterization signatures to 
confirm that this binding had taken place. That the imidazole/porphyrin binding be 



113 



persistent in the films is less important than that the imidazole is present and available 
to bind. 

5.2 Solution Studies 

Ligand effects as well as chromophore orientation and aggregation are 
primarily reflected in the shape or shift of the Soret Band or Band V. Therefore, the 
behavior of this band was carefully monitored. In addition, the ratio of Bands V to VI 
could be observed resulting from a change in the ligand environment. "^ Observed 
changes in the Soret Band in a non-coordinating solvent solution of a potential ligand 
are partially due to polarizability of the ligand. High polarizability of the axial ligand 
will lead to less negative charge induced on the porphyrin through the metal, and the 
Soret will shift to lower energies. • ^^ 

As mentioned in Chapter 4, the MnP4 spectrum demonstrates a strong 
tendency to intramolecularly bind to the phosphonic acid head-groups on the alkyl 
chain substituents. Because the imidazole binding is crucial in the catalysis studies, it 
was important to consider the ability of imidazole to replace phosphonic acid ligands 
in these systems. According to Arasasingham, the displacement of a water or hydroxy 
ion by imidazole is favored,"'* however, the ability of imidazole to replace 
phosphonate ligands had no literature precedence. 

5.2.1 MnPQ and MnP4 with ImH ' "* ^ '' '^ - ' -^' ' 

To determine the electronic effects of binding the imidazole in the absence of 
potential phosphonate ligands, an imidazole containing no long phosphonic acid chain 
(ImH) was added to a solution of MnPO, where the concentration of the MnTPP was 
held constant and the ratio of ImH:MnTPP was increased. With MnPO, as an excess 



114 



of the imidazole was added, an obvious blue shift of about 20 nm to ca. 458 nm was 
observedinthe A,n,ax(Figure 5.3A). ■ \ . • • ,'-\ 



0.10 



0) 

ii 0.06 



-1 ■ r 




600 



Wavelength (nm) 



Figure 5.3: Solvent response of A) MnPO and B) MnP4 to ImH. 



Upon addition of ImH to MnP4 solutions in CH2CI2, the original red shoulder 
on the Soret Band disappeared and was followed by a blue shift of the A-^ax to 458 nm 



115 



(Figure 5.3B). This peak was in reasonable alignment with the peak observed in the 
MnPO solutions under the same conditions. Imidazole ligands are not reported to 
encourage aggregation effects; therefore, these results again imply that the blue shift 
represents imidazole binding. This blue shift corresponds to a similar shift reported by 
Bruice et. al in their solution studies of a Mn-porphyrin with an increasing 
concentration of an imidazole ligand. ' " 

At the high concentrations of ImH relative to porphyrin, the imidazole filled 
either one or both of the axial positions on the metallo-porphyrin. The imidazole, 
therefore, successfully displaced the phosphonic acid. Also, in the solutions 
containing excess ImH, Band Va, which was assigned to fi"ee-base porphyrin, is 
reduced. With the competition between imidazole and phosphonic acid for the axial 
position, the phosphonic acid is obviously less likely to demetallate the Mn-porphyrin. 
At lower imidazole concentrations, the blue shift was less obvious. However, the 
effective concentration of phosphonic acids on the MnP4 in the vicinity of the metal 
was very high, so the solution concentration of ImH had to be large to compensate. 

5.2.2. MnP4 and MnPO with ImODPA 

When ImODPA was added to MnPO in CHCI3, the blue shift indicating 
imidazole or phosphonic acid binding was not observed. Instead, the Soret Band 
shifted from 477 nm to 483 nm. This red shift in the Soret Band is due to bromide 
displacing chloride as the axial ligand. Bromide originates from the ImODPA, which 
is originally a protonated amine-bromide salt. Mn-porphyrins with bromide ligands 
absorb at higher wavelengths than with chloride ligands because bromide is less 
electronegative and more polarizable than chloride shifting the Soret Band to lower 
energies.' 15 a close look at the region ca. 41 8 nm shows that Band Va is also 
appearing. This band has been attributed to the presence of free-base porphyrins. 



116 



which again, exist due to an irreversible binding of the phosphonic acid to the 
manganese central metal under certain conditions causing the demetallation of the Mn- 
porphyrin. 



0.06 



0.04 



o 0.02 



o 



3 



0.00 



0.03 

8 

c 

5 0.02 

o 

< 



0.01 - 



0.00 




350 



400 450 500 

Wavelength (nm) 



550 



Figure 5.4: Solvent response of A) MnPO and B) MnP4 to ImODPA. Legends 
indicate the molar ratio of MnP to ImODPA. 



ImODPA was also added to a CHCI3 solution of MnP4. The spectral behavior 
of this solution was more unusual. With 1 eq. of ImODPA, the red shoulder appeared 



117 



at 483 nm and the blue shoulder at 463 run was reduced in absorbance intensity. This 
spectral behavior corresponds to some of the phosphonic acid ligands being displaced 
by bromide. However, the bromide ligation did not appear to be quantitative, as was 
the case with the MnPO. After the addition of 10 eq. of the ImODPA, the intensity of 
the peak at 483 nm leveled off, and the 463 nm peak was at a minimum. With 20 or 
more equivalents of ImODPA, the blue Soret peak, or Band Vi, shifted to 460 nm and 
increased in intensity (Figure 5.4). This blue shift may indicate imidazole binding. 
Since the ImODPA amphiphile is terminated by a phosphonic acid, it would 
seem, again, that addition of this amphiphile would only serve to increase phosphonic 
acid binding. The competing ligands, however, make definitive identification of the 
peak at 460 nm difficult. Above a certain ImODPA concentration, it appeared that the 
imidazoles were displacing the remaining phosphonic acids and possibly some 
bromides and shifting Band Vi to higher energies. Therefore, a MnP(Br)(Im) species 
was probably present in solution at high ImODPA concentrations. In the films of the 
Mn-porphyrin, this competition may be avoided if the phosphonic acids are bound to 
the zirconated ODPA template. 

5.3 Film Studies 

5.3.1 . Langmuir-Blodgett Films containing substituted MnP4 

5.3.1.1. MnP4 substituted into an HDP A LB film . To understand the ability of 
the porphyrin to bind to the zirconium network in a preformed bilayer, a film was 
prepared by substituting a MnP4 layer into a LB film of hexadecylphosphonic acid 
(HDP A). By substituting the porphyrin film into an aliphatic capping layer, the hope 
was that the phosphonic acids could secure the porphyrin to the surface while the 
HDPA would be able to prevent porphyrin aggregation. The HDPA film formed in 



118 



this manner was well-organized and represented the most compact, and therefore most 
challenging, bilayer for MnP4 substitution. 

With the potential phosphonic acid binding sites in the Zr-network occupied 
with HDP A, the phosphonic acid tethers on the porphyrin were available for 
intramolecular ligation. From solution results, however, chloride or bromide could 
displace these phosphonic acid axial ligands (Figure 5.4). When the substituted film 
was studied by UV-vis immediately after the MnP4 SA process was complete, the A,,,,^ 
appeared at 454 nm. At such a high energy, the Soret Band implied that the 
chromophores were aggregating. Two lower energy shoulders could also be 
identified in the broadened Soret Band. The first appeared at ca. 463 nm, which may 
represent non-aggregated porphyrin domains, and the second appeared at ca. 478 nm at 
lower intensity indicating the presence of MnTPP(Cl) moieties. 

The film stability was examined by exposure to hot CHCI3 and a loss in 
absorbance intensity and a significant peak broadening was observed. After the first 5 
min in hot CHCI3, no ftirther decrease in the overall absorbance intensity of the film 
was perceived, and this spectrum represented only the chemisorbed porphyrins. With 
the physisorbed chromophores removed, the original red shoulder at 478 nm, which 
was also observed previously in the solution and films of pure MnP4, was more 
distinct, and there appeared to be two peaks. Bands Vi and Vii. From the two apparent 
Soret Bands, it seems that the remaining porphyrins were either MnTPP(Cl)(PA) (peak 
at 463 nm) or MnTPP(Cl) (peak at 478 nm) (Figure 5.5). 

Due to the splitting and broadening of the Soret Band, it is impossible to 
attribute a change in absorbance intensity just to removal of physisorbed 
chromophores. Some physisorbed chromophores were indeed removed, which was 
found by calculating the differences in peak areas before and after rinsing. Figure 5.5 
also reveals a decrease in the absorbance intensity ratio of Band V to Band VI. A 



119 



change in the ratio of Band V to Band VI is often reportedly associated with a change 
in the ligand environment^^-^i Unfortunately, many possible ligand changes are 
present in these systems. Most likely, as the film was rinsed, the chloride ligands 
could have displaced the phosphonic acid ligands. However, from the solvent studies 
of the pure MnPO with ethylphosphonic acid described in Chapter 4, a decrease in 
VA^I was associated with increased phosphonic acid binding. Therefore, trends in 
these ratios were difficult to follow because so many scenarios were possible. 



0.015 



0.010 



0.005 



0.000 



After SA 




350 400 450 500 550 

Wavelength (nm) 



600 



Figure 5.5: UV-vis of ODPA/Zr/HDP A, SA MnP4 film rinsed in hot CHCl, 



5.3.1.2. MnP4 substituted into an ImODPA/HDPA LB film. Though HDPA 
was easily transferred onto the Zr-ODPA template using the LB method, imidazole 
was much more difficult to deposit in this manner. The hydrophilic nature of both the 
phosphonic acid and the imidazole made this molecule unsuitable for Langmuir 



120 



monolayer formation. Therefore, the imidazole was diluted with HDPA or ODPA to 
avoid this problem. After the LB transfer of the ImODPA:HDPA mixtures, the MnP4 
was substituted into this layer. When films containing solutions ranging from 10% to 
75% ImODPA in HDPA were prepared and the MnP4 was substituted into these films, 
a pattern was observed (Figure 5.6). In the films of 10% ImODPA, SA MnP4, the 
ratio of VA^a was clearly less than 1 .0. This ratio increased with an increased 
concentration of ImODPA. Band Va, associated with the free-base porphyrin, drops in 
intensity with the addition of an imidazole ligand to compete with the phosphonic 
acid, as was seen in solution studies with MnP4 and ImH (Figure 5.3B). The 
presence of free-base porphyrin is possibly due to the irreversible binding of 
phosphonic acid to manganese that then pulls the manganese out of the center of the 
chromophore. Therefore, if the imidazole prevents the phosphonic acid from binding, 
it may preserve the integrity of the metallo-porphyrin. 

Likewise, though less dramatic, the ratio of VA'^I also appeared to increase 
with added ImODPA. Unfortunately, the ratio of VA^I in these films was not always 
reproducible or controllable, and this ratio fluctuation was observed in films when 
imidazole was absent. These inconsistencies could be due to the inability to 
completely direct the axial binding of the metallo-porphyrins. 

The Soret Band in Figure 5.6 demonstrates splitting after rinsing in hot CHCI3. 
The blue side of the peak occurs at 462 nm, which is slightly red shifted from the peak 
previously attributed to imidazole binding. However, as the percentage of imidazole 
increases, Band Vi appears to shift slightly to the blue. Also, the occurrence of the 
second, red band at 477 nm indicates the presence of halide axial ligands assigned to 
MnTPP(Cl). Due to the steric constraints from the film environment, it is unlikely that 
these porphyrins would form bis-imidazole complexes [MnTPP(Im)2]. Therefore, this 



121 



peak splitting has been assigned to the formation of asymmetric axially bound 
porphyrins coexisting with domains of MnTPP(Cl). 



0.04 



0.03 
CO 0.02 

i 

to 

< 0.01 



0.00 



■ 


Va 






7*5% imiri 




/ '• 







- - 25% imid 


, 


.* 






. . . in% imirl 


VI 


. ^ 


V 




* • • lU /o UMIU 






• v ' 


. 




. V" ^ 


/ ^ 




1 




.V ■"- 




f^^\ 


i 






'J\ 


y 


\ I 
\ \ 

\v 


- 


' 




^w/ 


\ \ 


■ 


' 






v^ 


■ * *^JZJ" *** ^ - 




1 


1 1 


.1 


^^^^^^^^ 



350 400 450 500 550 

Wavelength (nm) 



600 



Figure 5.6: MnP4 substituted onto ImODPA:HDPA LB films after CHCI3 rinsing. 



Rinsing in hot CHCI3 successfiilly eliminated any physisorbed chromophores. 
However, hot CHCI3 had a tendency to also cause a change in the ligand environment. 
Therefore, room temperature CHCI3 was examined in order to eliminate the 
physisorbed chromophores. After rinsing a 25% ImODPA/HDPA LB film, MnP4 SA 
film in room temperature CHCI3, the UV-vis showed behavior different compared to 
hot CHCI3. The absorbance intensity of the Soret Band decreased along with a shift 
from 456 rmi to 458 nm (Figure 5.7). 



■ff \ 



•'•.f i 



i y 



122 



0) 

o 

c 
nj 

o 

Si 



1 hr in RT CHCI, 




350 400 450 500 550 

Wavelength (nm) 



600 



650 



Figure 5.7: MnP4 substituted onto a 25% ImODPA/HDPA film, rinsed in room 
temperatvire and hot CHCI3. 



When the film was then exposed to hot CHCI3, the absorbance intensity 
decreased further and the red component of the Soret Band appeared at 477 nm. 
Therefore, physisorbed chromophores can be eliminated from the surface by using 
room temperature or hot solvents, but a change in the axial environment was only 
associated with rinsing in hot solvents. The band representing halide binding at ca. 
477 nm was not permanent and relaxed back to the original peak at ca. 460 nm after 24 
hr (Figure 5.8). 



* .-^ 



m 



8 0.015- 

c 



0.000 



overnight 




After MnP4 
substitution 
and rinsing 



400 500 

Wavelength (nm) 



600 



Figure 5.8: UV-vis of an ImODPA/ MnP4 film after drying. 



5.3.1.3. MnPO linked to ImODPA containing LB films. MnPO, a porphyrin 
with no alkylphosphonic acid chains, was assembled into ImODPA SA films to study 
the imidazole binding in a simplified system. The only available methods of 
absorbing MnPO to these films were through the imidazole or through aggregation of 
chromophores, i.e. physisorption, to already chemisorbed porphyrins. Upon rinsing, 
any physisorbed MnPO's should be washed away. Figure 5.9 shows the UV-vis of a 
layer of MnPO attached to a 25% ImODPA/HDPA LB film. The absorbance intensity 
of the MnPO film was much less than films containing the MnP4, but the porphyrin 
was clearly present indicating the imidazole binding was active. After 60 min in a hot 
CHCI3 solution, the absorbance intensity of the film did decrease indicating 
chromophore loss. However, the imidazole binding was somewhat stable to hot 
CHCIj^ as the chromophores were not completely removed after 60 min in the solvent. 



124 



The ratio of ImODPA to HDPA has been varied, and MnPO films were still 
successfully formed. Further, using an ImODPA/ODPA mixture instead of 
ImODPA/HDPA also worked to prepare the MnPO films. The 25% ImODPA films 
were studied primarily because of the balance of high imidazole loading and stable 
film behavior. HDPA was the primary diluent used because its chain length was 
appropriate to form a film with the imidazole group fully exposed and available for 
binding. 



g 0.003 

c 

(0 

■e 0.002 
o 

< 



0.000 



-0.001 



/■-/.u^ 



After MnPO 
attachment 




550 400 450 500 550 

Wavelength (nm) 



600 



Figure 5.9: MnPO attached to a 25% ImODPA/HDPA LB film and rinsed in hot 
CHCU. 



5.3.2. Mn-porphyrins substituted into self-assembled films of ImODPA 

5.3.2.1. UV-vis response of MnP4 film to /-butyl amine and CHCU rinsing. 
MnP4 was successfully substituted into a SA layer of ImODPA. After substitution, 
the films were rinsed with a solution of ^butyl amine (/-BuNHj) to deprotonate the 
imidazole and facilitate its binding to the porphyrin. The base /-BuNHj was chosen 



125 



for its bulky nature, which should prevent its binding to the porphyrin. Rinsing the 
films in /-BuNHj caused a blue shift in the Soret Band from 461 nm to 458 nm, which 
was therefore, assigned to imidazole binding to the central metal. It was unclear if 
imidazole binding was encouraged by deprotonation or if the elimination of some 
physisorbed and therefore, non-imidazole bound porphyrins made it easier to observe 
the imidazole binding. Because of the film structure, the complex is most likely a 6- 
coordinate MnTPP(Cl)(Im) system. 

After rinsing with hot CHCI3, the Soret absorbance intensity decreased 
significantly. Band Vii shifted to 477 nm, which was again associated with the 
formation of the five-coordinate MnTPP(Cl). Twenty-four hours after rinsing, the 
halide-binding peak at 477 nm was no longer detected. Instead, the band at 465 nm 
increases in intensity as the films were left to reorganize overnight as was observed in 
the pure MnP4 films. From the energy of the band at 465 nm, which is at lower 
energies than the Soret Band observed with imidazole binding, this peak was likely 
due to the binding of phosphonic acid ligands and not imidazoles (Figure 5.10). The 
Soret Band absorbance in this film was comparable to that of previous examples of 
pure MnP4 substituted imidazole films, implying roughly the same number of 
chromophores were present. 



.V 






1 



126 



0.03- 



SA MnP4 



rinse fBuNHj 

-•— rinse CHCUhot) . 




400 450 500 

Wavelength (nm) 



Figure 5.10: MnP4 substituted onto a pure ImODPA SA film. 



5.3.2.2. Reversibility of the chloride binding in MnP4 substituted films. The 
same reversibility of the halide binding was observed in the films of MnP4 substituted 
into SA ImODPA films. As shown in Figure 5.11, after 24 hr, the peak at 477 nm 
reduced in intensity while the peak at 460 nm increased in intensity. The halide 
binding could be achieved again by rewetting this film, and the cycle continued for up 
to a week. However, with each cycle, the halide binding became slightly more 
persistent. The reversibility of the binding, as observed by the shifts in the Soret 
Bands, indicated that the molecular behavior in these ultrathin films was not static. In 
fact, in thin films, the porphyrins were keenly sensitive to the environment, especially 
environments containing potential axial ligands. 



0.03 



0.00- 



425 



450 



127 



■ hot CHCI3 
■overnight 




475 



500 



525 



Wavelength (nm) 



Figure 5.1 1: Reversibility of the chloride/phosphonic acid binding. 



-• 5.3.2.3. Rinsing the substituted MnP4 films in chloride ion solutions . Chloride 
ions were deliberately added to the system by rinsing the films in a solution of tert- 
butylammonium chloride (r-BuNHj^ CI"). These results confirmed that the peak at 477 
nm in the films represented the MnTPP(Cl). When the MnP4 substituted film was 
rinsed in a room temperature EtOH solution of chloride ions, there was a clear Soret 
peak splitting with the red peak occurring at 477 nm indicating that in excess, chloride 
can bind at room temperature. When this film was then placed in a room temperature 
chloride solution in 9/1 EtOH/H20 (the SA solvent mixture), Band Vii disappeared 
and Band Vi intensified (Figure 5.12). 



128 



0.016 



0.012 



S 



TO 0.008- 
o 

CO 

§ 0.004 H 



0.000 H 

0.016 
0.012 



0) 

o 

c 

5 0.008 

o 
(/} 

< 0.004 



0.000 



MnP4 substituted 
RT CI- (EtOH) 
RT CI- (EtOH/HjO) 
A, CI-, (EtOH/HjO) 
overnight 




'w fixy 



ai»Jim*C^ 



■A— overnight (above) 

t-BuNH^ 
T— overnight 




450 500 

Wavelength (nm) 



550 



Figure 5.12: MnP4 substituted film rinsed in chloride and t-butylamine solutions. 



To fiirther promote chloride binding, the chloride-containing EtOH/H20 
solutions were heated. Heating the solution worked remarkably well at incorporating 
chloride binding, and the low energy peak developed significantly. After being left 
overnight, the red peak was much more persistent. This result may indicate that in the 



129 



presence of a large excess of chloride ions, chloride binding is much less mobile and 
more difficult to displace (Figure 5.12). 

Rinsing in /-BuNHj in room temperature EtOH caused the complete and 
immediate displacement of the chloride ligands. It is known that imidazoles tend to 
bind more efficiently to the porphyrins than chlorides, and so it is possible that the 
band at 462 nm represents imidazole binding." ' 

5.3.2.4. Preparing the substituted film from a chloride containing solution . To 
avoid phosphonic acid binding upon film formation, a layer of MnP4 was assembled 
from a 0.1 M solution of chloride onto an ImODPA layer (Figure 5.13). Initially, the 
chloride binding was not obvious, and the Soret was slightly blue shifted relative to 
phosphonic acid binding. 



0.020 



0.015 



0.010 



0.005 



0.000 



-l ■ 1 •- 



MnPO with 

ImH binding 

in CHCI 



SAMnP4 




350 400 450 500 

Wavelength (nm) 



550 600 



Figure 5.13: MnP4 substituted from a 0.1 M CI" solution onto an ImODPA layer, and 
compared to a MnPO solution with ImH binding. , . 



- , /• : 






^ » ,. 130 



After sitting overnight, the red chloride peak again emerged. This behavior is 
contradictory to results obtained previously where the chloride peak seemed to 
disappear with time. By adding a sufficient amount of chloride to fill the porphyrin 
axial positions, the phosphonic acids may be forced to bind to the zirconium network. 
The chloride axial binding is in equilibrium while the zirconium-phosphonate binding 
is not. Once the phosphonates are unavailable for ligation, the imidazole binding, 
which is evidenced by a blue shifted Band Vi, is more likely to be observed. Further 
proof of imidazole binding is in Band Va, which was earlier associated with 
phosphonic acid binding causing the demetallation of the Mn-porphyrin. This band 
was clearly absent in these films. 

5.3.3. Self-assembling the MnP4 and ImODPA from a mixed solution 

As an alternative method, a self-assembly solution of a 70/30 mixture of the 
ImODPA and MnP4 was prepared. The UV-vis of this mixed SA solution showed 
Band Vi at 460 and Band Vii at 477 (Figure 5.14). Band Vi in solution may represent 
phosphonic acid binding; however, as the film was formed these phosphonic acids 
were attracted to the zirconium network leaving the manganese available for binding 
the imidazole. The blue shift may also demonstrate some contribution from porphyrin 
aggregation. Rinsing with a hot solvent was consistently done to remove any 
physisorbed chromophores from these films, though in some cases, as shown in Figure 
5.14, there often was very little loss in absorbance intensity overall indicating that 
there were few physisorbed chromophores present initially. 



131 





0.05 
0.04 


after SA A- . - 
\/;\a 


SA solution 


0) 

o 


0.03 


-.•■■. / ■ V 


\ 






\ after 15 min 


c 

TO 

JD 

i- 

O 
(/) 

< 


0.02 




l/hotCHCIj 


0.01 


•.\ /; 


1 ; 




0.00 


- 


V__<>-^=s, 




-0.01 


1 


1 



400 500 

Wavelength (nm) 



600 



Figure 5.14: ImODPA/MnP4 self-assembled from 70/30 mixture and rinsed in hot 
CHCU. 



5.3.4. Other methods for preparing ImODPA and MnP4 containing films . 

5.3.4. 1. LB deposition of MnP4 followed bv substitution of ImODPA . Films 
of MnP4 were transferred from a water subphase at 5, 10, and 15 mN m"' (before, 
during and after the plateau region of the isotherm shown in Figure 4.7). ImODPA 
was substituted into these films from an EtOH/H20 solution. It was thought that, if 
the MnP4 could form a complete monolayer with phosphonic acids bound to the 
zirconium network prior to imidazole being present, it may be more likely that there 
would be imidazole binding instead of phosphonic acid binding the central metal. The 
UV-vis behavior of these films after the substitution of ImODPA and after rinsing 
with CH2CI2 at room temperature is shown in Figure 5.15. Unfortunately, there are 



132 



many complications with the films prepared in this way. First, the imidazole will most 
likely want to bind to the exposed surface of the porphyrin first; however, the 
phosphonic acid will be trying to bury into the film to find the zirconium network. 
Further, it is very difficult to confirm the presence of the imidazole in these films by 
any means. Therefore, this method was not rigorously pursued. 



350 



SA ImODPA 



Rinse RT MeCI^ 




400 450 500 550 

Wavelength (nm) 



600 



Figure 5.15: ImODPA substituted into a MnP4 LB film transferred at 10 mN m"'. 



5.3.4.2. LB transfer of mixed MnP4/ImODPA film at high dH. For 
comparison, a LB film was prepared from a spreading solution containing MnP4 and 
ImODPA in a 1 to 3 ratio. Ideally, a monolayer richer in ImODPA would have been 
used, but again, this is a poor amphiphile for LB films, and so this was impossible. 
These films were transferred at a variety of pressures and pHs. In a film transferred 
before the onset of the isotherm at a pH of 1 1.3 (Figure 5.16), the Soret Band after was 
at 459 nm. When the film was transferred at a MMA associated with little 



133 



chromophore aggregation, the Soret Band at 459 nm possibly corresponds to 
imidazole binding. The basic subphase was intended to deprotonate the imidazole 
and encourage its binding to the porphyrin while reducing the porphyrin aggregation, 
as seen in the case of the Pd-porphyrins in Chapter 3. Additionally, the basic subphase 
would deprotonate the phosphonic acids causing them to be even more hydrophilic - 
and more likely to be on the water surface. Unfortunately, it is very difficult to 
confirm the presence or binding of the imidazole in these films. XPS scans of the Nj^ 
region were not helpful in determining if there were two nitrogen environments 
present; therefore, these films were not a major focus for the preparation of catalytic 
films. 




400 450 500 550 

Wavelength (nm) 



600 



Figure 5.16: LB film of MnP4/ImODPA transferred from a 25/75 mixture on an 
aqueous subphase, pH 11 .3. a 



i ' 



J^ 



5 -*•« I 



?■' • 



134 






^'-v -^ 



5.3.5. Characterization of films containing MnP4 and ImQDPA bv XPS and ATR-IR 
5.3.5.1. XPS results of pure MnP4 and pure ImODPA films . Though the UV- 
vis is the most well documented method of characterization for porphyrin films, the 
ImODPA had no clear transition in the UV-vis. Therefore, to characterize the 
ImODPA self-assembled film prior to the MnP4 substitution, ATR-IR and XPS were 
employed. To study the formation of the ImODPA films by XPS, a layer of ImODPA 
was self-assembled onto a zirconated ODPA template on a silicon wafer. 





Rnn 


v> 








c 




D 




8 


600 


c 




^ 




<D 


400 


(U 




■D 




N 


200 


ra 




E 





(A 


1000- 


C 




3 

o 


800- 


o 




c 


600- 










0) 


400 ^ 


0) 




■a 


200 J 


0) 




N 






0- 


ro 




F 




1— 
o 


-200- 



-400 




410 



405 400 395 

Binding energy (eV) 



390 



Figure 5.17: XPS multiplex scan over the N,, region of A) ImODPA, and B) MnP4 
self-assembled films. The dashed line represents the Gaussian peak tit. 



U5 



The layer of pure imidazole was scanned by the multiplexing technique over the 
nitrogen peak (N,J, between 410 and 390 eV. The nitrogen peak, which could only be 
due to the presence of imidazole, was clearly a single Gaussian peak centered at 401.0 
eV with a FWHM of 2 eV (Figure 5.1 7 A). 

When a capping layer of pure MnP4 was self-assembled onto the zirconated 
ODPA template, a nitrogen peak was also seen (Figure 5.17B). However, in these 
films, the nitrogen peak was broadened to 3 eV and shifted to 399.0 eV. The nitrogen 
atoms of the imidazole and the porphyrin are in two different envirormients, and a 
small change in their binding energies would be expected. Therefore, in the mixed 
film, two peaks should be present in the N,s region, one representing the imidazole and 
one representing the porphyrin. 

5.3.5.2. XPS results from mixed self-assembled films . A film SA from a 
70/30 mixture of ImODPA and MnP4 in CH2CI2 was studied by XPS to determine if 
two nitrogen environments could be detected. There was a strong peak at about 402.0 
eV and a distinct shoulder or second peak of weaker intensity at 399.0 eV indicating 
the presence of both nitrogen environments, and hence of both the imidazole and 
porphyrin within the films (Figure 5.18). Alternatively, a mixture of the porphyrin and 
imidazole in a similar ratio was self-assembled from 9/1 EtOH/HjO. The nitrogen 
region of these self-assembled films showed two peaks, the first centered at 402.0 eV 
and the second, a broader peak of approximately equivalent intensity centered at ca. 
400.0 eV (Figure 5.19). These peaks align reasonably well with those of the XPS 
studies of the pure imidazole (40 1 .0 eV) and porphyrin (399.0 eV). 

Again, these results indicate the presence of both the imidazole and the 
porphyrin in these self-assembled films. Similar results were observed by Offord et al. 
in films of ruthenium or osmium metallo-porphyrins that were adhered to a SA thiol- 



136 



on-gold surface.30 XPS, therefore, clearly demonstrated the presence of both the 
porphyrin and the imidazole in the films. 




410 



405 400 395 

Binding Energy (eV) 



390 



■■■. .•■ *i 



Figure 5.18: XPS multiplex scan of Nls region of ImODPA/MnP4 film self- 
assembled out of 70/30 CH2CI2 solution. The dashed lines represent the Gaussian 
peak fits. 




410 405 400 395 

Binding Energy (eV) 



390 



Figure 5.19: XPS multiplex scan of ImODPA/MnP4 film self-assembled from a 70/30 
mixture in EtOH/HjO. The dashed and dotted lines represent gaussian peak fits. 



137 



5.3.5.3. ATR-IR characterization of the films containing substituted MnP4 on 
a self-assembled ImODPA layer. An ODPA monolayer transferred by the LB 
technique onto a Zr-ODPA template leads to alkyl peaks at 2918 cm-i and 2850 cm-i 
corresponding to the Va(CH2) and Vs(CH2) stretches, respectively. For alkyl chains in 
a majority /ra«5-configuration with crystalline order, the Va(CH2) peak usually falls 
between 2918 cm"' and 2920 cm-'. A shift to higher energies indicates the 
introduction of disorganization within the hydrophobic region of the film. In addifion, 
the full width at half maximum (FWHM) of the Va(CH2) peak can be a measure of the 
packing and conformational order within the alkyl region. An organized, close-packed 
film has a FWHM of approximately 17 cm"', which can stretch to 35 cm*' upon 
disorganization of the film. 

In films formed by both the LB and SA of ODPA fi-om a 9/1 EtOH/HjO 
solution onto a zirconated ODPA template, the Va(CH2) stretch comes at 2918 cm-' 
and the Vs(CH2) is at 2850 cm-'. Therefore, these films are mostly trans and close- 
packed. In addition, the absorbance intensity of the Va(CH2) peak in the capping layer 
is between 0.012 and 0.015 au. An alkyl absorbance intensity of this magnitude is 
therefore associated with the formation of a complete, organized monolayer of an 
octadecyl amphiphile. 

In order to further confirm the presence of imidazole in the films formed by the 
SA of ImODPA onto a zirconated ODPA template, the Va(CH2) and Vs(CH2) peaks 
were monitored and compared to the ATR-IR results obtained from the ODPA 
monolayer. The ATR-IR spectra of the ImODPA films were referenced to the 
zirconated ODPA template; therefore, what was plotted were the results from just the 
ImODPA capping layer. When the Zr-ODPA template had soaked in a 9/1 EtOH/H.O 



\- \: ■ :■ ' -i^W-^ 



138 



solution of ImODPA for five minutes, the absorbance intensity of the Va(CH2) peak 
was 0.004 au, and the Va(CH2) and Vs(CH2) peaks appeared at 2926 cm-' and 
2854 cm"i, respectively. After 20 min, the absorbance increased to 0.0065 au and the 
peaks shifted to 2920 cm"' and 2850 cm"'. There was no further shift observed in 
these peaks; however, the absorbance slowly increased over 2 hr, at which time the 
absorbance intensity remained constant at 0.012 au. These results proved the presence 
of a monolayer of ImODPA (Figure 5.20 and 5.21). However, the FWHM of 28 cm'' 
was significantly greater than those seen in ODPA monolayers, indicating a level of 
disorganization in these self-assembled films. A peak was also observed at 1307 cm'', 
which corresponds to the presence of C=N stretches from the imidazole. Hence, ATR- 
IR studies confirm the presence of a monolayer of ImODPA formed from the SA of 
this amphiphile fi"om EtOH/HjO. 



0.015 



0.010 



0.005 



0.000- 



-0.005 



v(CH,) = 2918cm' 



3000 



#■ 



x5 



CHj deformation 
1464 cm'' 




2750'' 1500 

\Afevenumbers (cm"^) 



1250 



Figure 5.20: ATR-IR of ImODPA SA film. 



139 



0.014 
0.012 
0.010 

o 0.008 

c 

■Q 0.006 H 
o 

^ 0.004 
< 

0.002 
0.000-1 




100 150 200 

SAtime (min) 



250 



Figure 5.21: Increase in absorbance intensity of 2918 cm"' peak in ImODPA with SA 
time. 



After a layer of pure ImODPA was formed, a layer of MnP4 was self- 
assembled to monitor the incorporation of the porphyrin into these films. After 1 hr, 
, the absorbance intensity of the Va(CH2) peak leveled off at 0.016 au (as referenced to 
the ImODPA monolayer), but the peak consistently appeared at 2924 cm-'. With such 
a large chromophore on the exterior of the hydrophobic region, the alkyl groups would 
be expected to be more dispersed and hence, show significantly lower absorbance 
intensity. Also, the addition of these bulky amphiphiles would be expected to increase 
the disorganization of the monolayer and shift the alkyl peaks to higher energies. 
However, the absorbance intensity implies a relatively high density of alkyl chains 
fi-om the MnP4 units with only slightly more disorganization than seen in SA ODPA 
films. The physisorption of the MnP4s could result in the alkyl chains burying into the 
hydrophobic region of the chemisorbed film, leading to a high density of alkyl chains 
along with some semblance of organization within this region. The possible 



140 



physisorption of a number of chromophores is expected due to the propensity of the 
porphyrins to aggregate. However, this impHes a significant reorganization of the 
underlying imidazole monolayer in order to have room to incorporate these alkyl 
chains. 

After 30 min in hot CHCI3, the absorbance intensity leveled off at 0.01 1 au and 
Va(CH2) peak occurred at 2928 cm-', as referenced to the imidazole monolayer (Figure 
5.22A). This result indicates that the true porphyrin containing monolayer contains 
alkyl chains that are disorganized and include a number of gauche interactions. After 
removing the physisorbed chromophores, the true MnP4 monolayer characteristics 
were better defined. 

Also, the SA of this film for only 5 min, leading to an absorption intensity of 
0.004 au, formed a SA film containing approximately 30% of a complete imidazole 
monolayer. This incomplete film was then exposed to a MnP4 solution. After 30 min, 
the absorbance intensity leveled off at 0.018 au and the Va(CH2) was at 2926 cm-'. 
After only 10 min in hot CHCI3, the absorbance dropped drastically to 0.005 au and 
Va(CH2) shifted to 2937 cm"' (as referenced to the imidazole monolayer). However, 
the IR shows that the SA of a 30% ImODPA layer resulted in a relatively poor 
monolayer, which left many defect sites. Originally, it was thought that these defect 
sites would allow for the more straightforward inclusion of the porphyrin. 
Unfortunately, the imidazole layer in such a state may be easily removed fi-om the 
zirconated ODPA template, and the final film may include less imidazole and 
porphyrin (Figure 5.16B), and therefore, have a lower overall alkyl absorbance in the 
IR. 



141 




3100 30'00 2900 2800 



Wavenumber (cm"^) 



3000 2900 2800 2700 
Wavenumber (cm"^) 



Figure 5.22: ATR-IR of alkyl region of: A) MnP4 substituted on a 100% ImODPA 
base capping layer, B) MnP4 substituted on a 25% ImODPA base capping layer. 



5.4 Conclusions 



Thin films containing both the MnP4 and the imidazole on the surface of a 
zirconium phosphonate film have been successfully prepared by a variety of methods. 
The imidazole is present in the films for the purpose of acting as a heterocyclic ligand 
to promote the catalytic behavior of the MnP4 chromophore. Because of the 
flexibility of assembling the zirconium phosphonate thin films, the procedure for 
preparing these films has been varied to optimize the incorporation of ImODPA and 
MnP4. Though aggregation is difficuk to identify due to the strong optical influences 



m 



of the different axial ligands, the evidence indicates that the aggregation is probably 
not significant in these films. 

Originally the spectral changes observed with rinsing were considered due to 
some chromophore reorganization or possible solvent effects. However, the peak at 
477 - 478 rmi aligns with the porphyrins' solvent UV-vis spectra only in the presence 
of added chloride or bromide; therefore, halide binding has been concluded to cause 
this red peak in the films as well. As mentioned, the chloride or bromide ligand 
causing the Soret Band change is probably associated with the MnP4 as the original 
axial ligand or bound to the imidazole through the HBr sah. When solubilized in hot 
solvents, the halides then have the opportunity and the preference for binding at the 
porphyrin metal center. The halide binding can be seen after rinsing in hot CHCI3 or 
hot CH3CN, as shown in Figures 4.1 1 and 4.14. 

From ATR-IR, XPS, and UV-vis, the existence of ImODPA and MnP4 within 
the films is confirmed. From the ATR-IR, the formation of the imidazole-containing 
capping layer was determined. In addition, UV-vis and XPS confirms the presence 
and behavior of the pure MnP4 capping layer. XPS results show a nitrogen 
environment within the porphyrin, which differed in binding energy fi-om the nitrogen 
peak for a pure imidazole film. In the mixed films, the ATR-IR demonstrated the 
addition of porphyrin alkyl and aryl environments upon substitution into the 
imidazole-capping layer. 

The ability of the MnP4 films to reversibly bind different axial ligands is 
inferred from the activity in the Soret region of the UV-vis spectra upon heat and 
solvation. It appears that the Soret Band splits into two peaks, which typically 
represent the binding of a chloride ligand and the binding of either a phosphonic acid 
or an imidazole ligand in the fifth and sixth coordination sites. Further, it is possible 
that porphyrin domains form with slightly different ligand environments. 



143 



From solvent studies, it is clear that the imidazole can displace either 
phosphonic acid or chloride ligands if present in a high enough concentration. This 
binding is in equilibrium, making this again, difficult to consistently observe in the 
UV-vis. However, it is believed that in the case of the catalysis reaction, the binding 
will be favored due to the excess of imidazole present and that this binding will cause 
the activation of the porphyrin for catalysis using peroxide oxidants even if it is not 
consistently observed by UV-vis. 



•^►- 



CHAPTER 6 
MANGANESE PORPHYRIN AND IMIDAZOLE CONTAINING 
ZIRCONIUM PHOSPHONATE THIN FILMS AS CATALYSTS 



6.1 Background 

Reactions using immobilized catalysts have recently gained much interest. First, the 
immobilized porphyrins are somewhat protected from destructive oxidation because they 
are ideally isolated from one another on the surface. Elimination of destructive oxidation 
can improve the catalyst's turnover rates and overall reaction yields. Secondly, vs^hen the 
catalyst is immobilized, its recovery is trivial. Therefore, if the catalyst is still active, the 
film can be recycled and used in another catalytic reaction. With these motivations, 
reactions using the MnP4 films, either pure or with ImODPA, were studied. Along with 
the overall reaction yield, the optical intensities of the porphyrin films before and after 
the reaction were monitored to determine if the porphyrins were bleached or removed 
from the surface during the course of the reaction. 

Battioni et al. examined the epoxidation of olefins using hydrogen peroxide as the 
oxidant.! 16 in this report, Battioni investigated conditions using 5,10,15,20-tetrakis(2,6- 
dichlorophenyl)porphyrin (MnTDCPP(Cl)) and 5,10,15,20- 

tetrakis(pentafluorophenyl)porphyrin (MnTFPP(Cl)) with both A) excess substrate and 
B) excess oxidant. The molar ratio in condition Awith MnTDCPP(Cl) was 800:10:20:1, 
cyclohexene to imidazole to H2O2 to catalyst, and under these conditions, reported 
cyclohexene oxide yields of 97% were observed. The molar ratio 



144 



M5 



investigated under condition B, 40:20:200:1 cyclohexene to imidazole to HjOj to 
catalyst, with the same porphyrin gave epoxide yields of ca. 91%. When the catalyst 
was MnTFPP(Cl), called MnPO in our studies, the yields were 58% under conditions 
A and 74% under conditions B. The porphyrins containing halide substitutents on the 
phenyl rings appeared to be fairly resistant to bleaching over the course of the 
homogeneous reactions. 

Baciocchi et al. studied the homogeneous epoxidation of cyclooctene to 
cyclooctene oxide using both manganese and iron porphyrins as catalysts, comparing 
the effects of electron donating vs. electron-withdrawing substituents on the 
tetraphenyl rings.99 The porphyrins studied included 5,10,15,20-tetrakis(2,6- 
dimethoxyphenyl)porphyrin (MnTDMeOPP(Cl)), the above MnTDCP(Cl), and 
5,10,1 5,20-tetraphenylporphyrin (MnTPP(Cl)). Results indicate that MnTPP(Cl), in a 
1 :40:200 ratio to substrate and hydrogen peroxide gave epoxide yields of 15% as 
referenced to the initial substrate concentration. The MnTDCPP(Cl) derivative in the 
same molar ratio to substrate and oxidant gave a 36% yield (as compared to 91% in 
the Battioni report), and the MnTDMeOPP(Cl) catalyst gave a 78% yield. 

When the oxidant was instead iodosylbenzene, or PhIO, the substrate was used 
in excess with a molar ratio of 1 :400:20, porphyrin to substrate to PhIO, respectively. 
The epoxide yields ranged from 73 to 80% with reference now to the oxidant. In both 
the PhIO and HjOj epoxidation reactions, the bleaching effects observed with the 
MnTDCPP(Cl) and MnTDMeOPP(Cl) catalysts were mild, under 5%. Only the 
unsubstituted MnTPP(Cl) showed up to 10% bleaching reported over the course of the 
2 hr reaction in the presence of PhIO. 

Thin films of catalytic porphyrins have also been studied. Abatti et al. 
investigated LB films containing an iron(III) 5, 10, 15, 20-tetrakis(tetradecyl-2-N- 
pyridyl) porphyrin. With the inclusion of alkyl chains on the four-pyridine rings, clear 



146 



hydrophobic and hydrophihc regions were defined in the molecule and this amphiphile 
formed monolayers on the water surface. These monolayers could be transferred onto 
hydrophobic substrates placing the chromophore on the exposed surface. The 
immobilization of these porphyrins saw a doubling of the epoxide yield as compared 
to the same catalyst to reactant ratios in solution with PhIO as the oxidant. 
Unfortunately, the transferred monolayer by itself was not stable and was removed 
almost instantly from the surface in room temperature CH3CN. Abatti et al. used a 
poly-vinylalcohol film to stabilize the porphyrin film for up to 4 hr.^^ 

As an alternative to the homogeneous reactions and coated LB films using 
porphyrins as catalysts, our work focused on immobilizing active porphyrins in 
independently stable film structures. As mentioned in Chapters 4 and 5, the process 
involved incorporating manganese(III) 5, 10, 15, 20-tetrakis(tetrafluorophenyl-4- 
octadecyloxyphosphonic acid)porphyrin (MnP4), with and without imidazole 
octadecylphosphonic acid (ImODPA), into zirconium phosphonate LB and SA films. 
The film structures were stable in hot CH2CI2 over 60 min and were expected to be 
stable in the catalysis reactions. 

The flow cells described and illustrated in Chapter 2 were employed to run 
reactions using the porphyrin containing films with or without imidazole. The 
different reaction mixtures were prepared and the blank and homogeneous solutions 
were separated into different vials, where they were stirred for a given length of time. 
The blank solution was pumped through a cell containing the film of interest for the 
same amount of time. In most cases, the reactions were run for 2, 6, or 24 hr. 

Because the extent of the reaction would be measured by GC, the first step in 
this study involved determining the GC sensitivity factors for the cyclooctene and 
cyclooctene oxide. A decane standard was used in PhIO reactions, and an o- 



147 



dichlorobenzene standard was used in the HjOj reactions. The sensitivity factors were 
calculated using Equation 2.2: 

kc^=^^ (2.2) 



and were found to be 1 .297 for cyclooctene oxide (CyOO) and 1 .008 for cyclooctene 
(CyCO) with decane. After average k values were obtained for the CyO and CyOO, 
the catalysis yields were determined from a GC of the reaction mixture using Equation 
2.3. The amount of product was calculated in grams, and referenced to the theoretical 
mass to obtain a percent yield. An empirical sensitivity factor for each compound is 
necessary due to the fact that the GC detector does not respond identically to each 
solute. 

kcyOO-^sA:yOO 

6.2 Results 

6.2. 1 Catalvsis with PhlO as the oxidant 

6.2.1.1. Time dependence of oxidation yields . A stock solution of CyO and 
PhIO with decane was prepared in CHjClj solution at a 1000: 125:500 molar ratio in 
CH2CI2. A 1 mL aliquot contained 40 ^mol CyO, 5 ^mol PhIO and 20 fimol of 
decane, which was then used for the blank reactions with no porphyrin present. 
Considering the amount of porphyrin in the films to be approximately 1x10"' moles, 
the molar ratio of the corresponding homogeneous reaction was .001 :40: 5:20 in, again, 
1 mL of CH2CI2. Also, ca. 1 mL of the stock solution was introduced into the flow 



148 



cells containing the MnP4 film. The films used in these experiments were prepared by 
self-assembling a monolayer of MnP4 onto a zirconated ODPA template and rinsing 
this template with hot CHjClj. The UV-vis band, Va, which has previously been 
assigned to the free-base porphyrin, is clearly present in the films used for the 24 hr 
and 2 hr experiments. 

When these reactions were allowed to proceed over 24 hours, GC results 
indicated that the average yield in the blank was ca. 1.5 %, in the homogeneous 
reaction was 8.4%, and in the films, between 20% and 35%. These percent yields, as 
calculated relative to the internal standard peak, were fairly consistent. 



0.05 



0.00 



After SA 

After rinsing CHCI3 

After Catalysis 




350 



400 



450 



500 



550 



600 



Wavelength (nm) 



Figure 6. 1 : SA MnP4 film before and after 24 hr. catalysis run with 40:5:20 
cyclooctene: PhIO:decane in CHjClj. 



Interesting to note is the fact that, even after 24 hr in the catalysis reaction, 
there does not appear to be a significant bleaching effect. The porphyrin appears to 
still be present and intact in the film. What is noticeable is the fact that the Soret Band 



#•' 



149 






has reverted back to its peak energy at 460 nm. From the known mechanism of this 
epoxidation reaction, the porphyrin loses its axial ligand and becomes oxidized; 
therefore, there is evidence for this ligand exchange, which is witnessed in the UV-vis 
of the films after the catalysis (Figure 6.1). 

When this same reaction was run for only 2 hr, the overall yields, as expected, 
decreased (Table 6.1). These films were prepared in the same manner, as were the 
films studied in the previous catalysis experiment. The ratio of Bands V:V1 changes 
slightly, as the axial ligand environment is not identical in both cases. Again, the UV- 
vis behavior before and after the catalysis reaction was studied and is shown in Figure 
6.2. 



0.02 



(D 
O 

c 

^ 0.01 

i_ 
o 

(0 

< 



0.00 



rinsing CHCI3 
After catalysis 




350 400 450 500 550 

Wavelength (nm) 



600 



Figure 6.2: SA MnP4 film before and after 2 hr. catalysis run with 40:5:20 
cyclooctene: PhIO:decane in CHjClj. 



150 



Considering the original UV-vis spectra of the films indicated in the 2 hr and 
24 hr experiments, there appeared to be a significant contribution fi-om fi-ee-base 
porphyrin in these films. From the studies described in Chapter 4 and 5, it was known 
that self-assembling the MnP4 from a solution containing chloride ions helped 
discourage the phosphonate from demetallating the porphyrin. To test if a decrease in 
the film loading of free-base porphyrin affected the catalytic results, a film SA from a 
chloride containing solution was employed for the 6 hr experiment. Again, some 
bleaching of the porphyrin was observed, but it did appear that there was a significant 
amount of the original chromophore present after catalysis. Also, there was no 
significant contribution from fi-ee-base porphyrin to the UV-vis spectra even after the 
catalysis run, so the catalysis does not cause the Mn-porphyrin to demetallate (Figure 
6.3). However, the epoxide yield was not dramatically improved by eliminating the 
free-base porphyrin from the films. 



V 

'<0 



o 

c 
03 

n 

i_ 

o 

CO 

< 



0.035 
0.030 
0.025 
0.020 
0.015 
0.010 
0.005 
0.000 



-0.005 



- 


1 ' 


■ ■ V • 


' 1 ' 


- 


- 


/I 


— 


— After SA 

- - rinsed CH^C!^ 

catalysis 


- 


/ \ 


|\ 


t 

1 
t 

1 
1 

» 




- 


' r • 


— 1 — ' — 


\ \ 


• 1 ' 


^ 



350 400 450 500 550 

Wavelength (nm) 



600 



Figure 6.3: UV-vis of MnP4 film SA from chloride containing solution used in 
catalysis with PhIO after 6 hr. 



151 



The behavior of the MnP4 films in the epoxidation reaction with PhIO over 6 
hr showed up to a 20% reduction in the absorbance intensity of the peak at 464 nm. 
The film could be recycled with ca. a 50% loss in catalytic activity, but catalysis could 
still be observed. 

For comparison, the bleaching of the porphyrin in the homogeneous solution 
was studied (Figure 6.4). It appears that after 6 hr., there was as much as a 50% 
bleaching of the active chromophores in solution. After 12 hr, the bleaching 
observed was about the same as at 6 hr. The porphyrins immobilized in the zirconium 
phosphonate films appear to be slightly more stable under most catalysis conditions. 



Table 6.1 : Time dependence of epoxidation of cyclooctene using 40 |amol 
cyclooctene and 5 |amol PhIO in ImL of solution. To the homogeneous reaction was 
added Inmol of MnPO. 





Experiment 


Cyclooctene oxide 
Yield 


Turnovers* 


Porphyrin 
Bleaching** 




24 hr 








« 


Blank 


4.98% 






V 


Homogeneous 


8.43% 


422 






Films 


35.1%, 20.8% 


1755, 1040 


<20% 




6hr 










Blank 


1.0%-1.5% 








Homogeneous 


6.0%-7.0% 


300 


< 50% 




Films 


30.3%, 29.0% 


1515,1450 


<20% 




2hr 










Blank 


0.5%-1.0% 








Homogeneous 


3.0%-3.4% 


150-170 ' 


= ' 




Films 


13.3%, 12.7% 


665, 635 


None 



* Turnovers calculated on number of oxidation cycles 

** Bleaching determined from decrease in intensity of Soret Band in UV-vis 



152 



From Table 6.1, it is clear that the conversion of CyO to CyOO is time 
dependent. The best yields were usually achieved after 24 hr; however, the difference 
between 6 hr and 24 hr was small in some cases. Although the absorbance intensity 
prior to the reaction was roughly the same as in the case of the SA films, after 24 hr, 
the overall absorbance intensity has decreased. This result indicates that a percentage 
of the chromophores have been bleached or removed from the surface. Considering 
studies done previously on the stability of the porphyrin containing zirconium 
phosphonate SA and LB films after rinsing in hot solvents, the chromophores are 
probably not removed but have likely been oxidized rendering them useless as 
catalysts. However, the bleaching process was not complete and there were still a 
number of catalytic porphyrins available on the surface. 



0.15 



0.10 



S 



(0 

43 0.05 



0.00 



MnP0(1 x10"'M) 
after 6 hr. cataysis 



T ' r 



MnP0 1 x10"*M 




350 400 450 500 550 600 

Wavelength (nm) 



Figure 6.4: Bleaching of MnPO in homogeneous catalysis reaction with PhlO. 



153 



■*,>■ : 



■ -P it V 



-0.005 



350 



400 



450 



-After LB 
rinse hot CHCI3 
24 hr. catalysis 




500 



550 



600 



Wavelength (nm) 



Figure 6.5: MnP4 LB film before and after 24 hr catalysis reaction. 



6.2.1.2. Oxidation yield dependence on film preparation method . The above 
described catalysis results were observed using MnP4 SA films. Alternatively, pure 
LB films of MnP4 were also used in the reaction cells. A MnP4 film was transferred 
by the LB technique at 20 mN m"' where the density of chromophores was high and 
aggregation was observed. The film was rinsed to remove any possibly physisorbed 
chromophores present due to the compressed nature of the film. These films were then 
used in the reaction flow cells over 24 hr with a 40:5:20 ratio of CyO to PhIO to 
decane. The corresponding blank and homogeneous reactions gave yields of, again, 
around 4% and 10% respectively; however, the yields with the films were around 
15%, which is slightly lower than that seen in the case of the self-assembled porphyrin 



\M 



films (Figure 6.5). The lower yield could be due to the aggregated nature of the 
chromophores reducing the active porphryin surface area. 

Table 6.2: Conversion of cyclooctene to cyclooctene oxide with 40 |umol cyclooctene 
and 5 (amol PhIO in ImL of solution using MnP4 LB film in 24 hr. 



LBMnP4 


Cyclooctene oxide 
Yield 


Turnovers 


Bleaching 


Blank 


4.0% 






Homogeneous 


11.0% 


550 


<50% 


Film 


15.0% 


750 


<20%* 



* The amount of bleaching was difficult to determine due to the broadened Soret 
Band observed prior to the catalysis reaction. 



6.2. 1 .3. Oxidation vield dependence on reactant ratios . Changing the ratio of 
substrate to oxidant was also investigated. Whereas the prior studies focused on a 
ratio of 40:5, 60:5 and 20:5 substrate to oxidant ratios were also examined over 24 hr. 
Changing the substrate to oxidant ratio resulted in yields similar to that seen for the 
40:5 experiment. In the 60:5 case, the overall yield was, again, around 20%); however, 
the porphyrin film experienced slightly less bleaching than observed previously with 
less substrate. The 20:5 case behaved similarly to the 60:5 and 40:5 experiments, and 
the yields were approximately the same (Table 6.3). 



155 



Table 6.3: Conversion of cyclooctene to cyclooctene oxide with varying cyclooctene 
to PhIO ratios in ImL of solution using MnP4 SA films over 24 hr. 



Experiment Cyclooctene Turnovers Bleaching 

Oxide Yield 



60:5:20 






Blank 


3% 




Homogeneous 


7% 


350 


Films 


20% 


1000 


20:5:20 






Blank 


2% 




Homogeneous 


5% 


250 


Films 


18% 


900 



<15% 



<15% 



6.2. 1 .4 Oxidation dependence on presence of heterocvclic lieand . Addition of 
an imidazole heterocyclic ligand is commonly reported to enhance the catalytic 
behavior of Mn-porphyrins in the presence of a peroxide oxidant. However, little was 
mentioned in the literature about the porphyrin response to this ligand when using PhIO 
as the oxidant. The imidazole was not only expected to encourage the heterolytic 
cleavage of the 0-0 bond of the peroxides but also to stabilize the oxidized manganese 
state; therefore, some extent of catalysis improvement might be expected. To examine 
this sensitivity, a SA ImODPA/SA MnP4 film was prepared and rinsed to remove 
physisorbed chromophores. Using a 40:5 substrate to PhIO ratio, the catalysis reaction 
was run over 24 hr (Table 6.4). 



156 



0.025 



0.020 



0.015 



-0.005 



After SA 
rinse hot CHCI3 
24 hr Catalysis 




350 400 450 500 550 

Wavelength (nm) 



600 



Figure 6.6: SA ImODPA/SA MnP4 studied with PhIO for epoxidation of cyclooctene. 



The CyOO yields using these mixed films were actually reduced relative to the 
pure porphyrin films with PhIO. The lower yield could be due to a decreased 
porphyrin concentration in the mixed films. The UV-vis spectrum of the films used 
for this study demonstrated that the overall absorbance of the Soret Band prior to 
catalysis was lower than that observed in the pure porphyrin film (Figure 6.6). This 
result corresponded to the behavior often observed in Chapters 5 where porphyrin 
substitution into a preformed capping layer resulted in a slightly lower loading than in 
the pure self-assembled films. 



157 



-♦;»;■ 



'^ 



Table 6.4: Conversion of cyclooctene to cyclooctene oxide with 40 }amol cyclooctene 
and 5 ^mol PhIO in ImL of solution and in films containing imidazole. 



SAImODPA/SA Cyclooctene oxide Turnovers Bleaching 

MnP4 Film Yield 



Blank 1.5%-2.5% 

Homogeneous 17.0%-20.0% 850-1000 

Film 15.0% 750 <20% 



6.2. 1 .5. Oxidation yield dependence on cell vs. vial used for reaction . The 
above results appeared encouraging, with the epoxide yield consistently greater in the 
immobilized catalyst reactions run in the flow cells over the homogeneous and blank 
reactions run in the vials. When blank solution mixtures were run through the 
catalysis cell with blank films, surprising results were obtained. After 6 hr, the blanks 
in the vials ranged in product yield fi'om 1.0% to 2.0%; however, in the flow-cells, the 
epoxide yield was 6.0%-6.5%. The discrepancy between the yield in the blank cell 
and in the vial was not related to a change in the overall concentration. However, the 
difference was probably due to either the loss of internal standard into the flow-cell 
material or the fact that the solution experiences more efficient stirring in the flow-cell 
than in the vial. 

Further, homogeneous reactions were also run in the flow-cell to examine if the 
observed increased product yield with the catalytic films was due to the immobilized 
catalyst or the act of performing the reaction in the cell. These results are shown in 
Table 6.5. From a series of experiments, including GCs of a solution run after the 



158 



PhIO reaction contents were removed from the cell, it appeared that decane is retained 
in the cell. The decane is even observed in the GC traces after the cell has been 
thoroughly cleaned, implying that this molecule has a tendency to adsorb onto the 
tubing and/or cell material and dissolves out slowly. The amount of decane seen after 
the cells were cleaned is small relative to the original peak, but this may be the source 
of some discrepancy observed between blanks and homogeneous reactions run in and 
out of the cell. 

Overall, it appears that there is an improvement observed in the catalytic 
activity of the porphyrin from the homogeneous solution to that seen in the 
inmiobilized films. These systems clearly need to be studied using an alternative 
internal standard to potentially eliminate the discrepancy observed with leeching of the 
decane into and then out of the flow cell. 



Table 6.5: Comparison of blanks and homogeneous epoxidation yields in vials vs. in 
the reaction cells. 



SAMnP4 




Cyclooctene oxide 
Yield 


Turnovers 


Blank in vial 




1.3%-2.0% 




Blank in cell 




6.0%-6.5% 




Homogeneous in 


vial 


6.0% 


300 


Homogeneous in 


cell 


9.0%-14.0% 


450-700 


Film in cell 




20.0%-21.0% 


1000 



159 



6.2.2. Catalysis using H oO, as the oxidant 

6.2.2.1. Oxidation with iC times less catalyst vs. other reactants (80 |j.mol 
H^Ot or 40 f^mol cyclooctene vs. 1 nmol MnP) . The above-described PhIO 
epoxidations were performed with Inmol of catalyst relative to 40 |j,mol of substrate 
and 5 jxmol of oxidant. The catalyst in these studies is ca. 10^ times lower in 
concentration than that suggested in the literature studies. The results, however, 
suggest that the catalyst is still active at this concentration under these conditions. 
Additionally, in the PhIO reactions, excess substrate was used to protect the catalyst 
from bleaching. As mentioned in section 6.1, with peroxide oxidants, both excess 
substrate and excess oxidant have been investigated. 



0.025 



0.020 



-0.005 



After SA 

rinse CHCI3 

24 hr. catalysis 




350 400 450 500 550 600 



Wavelength (nm) 



Figure 6.7: SA ImODPA/SA MnP4 studied in the epoxidation of cyclooctene using 



H2O2 



160 



When experiments similar to the Bacciochi study with excess substrate were 
reproduced using 400 fimol cyclooctene and 80 ^imol H2O2 in ImL of solution, with 
25 jimol imidazole and 1 nmol of MnP added to the homogeneous reaction, very little 
epoxide was detected in the blank or homogeneous reactions.^^ Fortunately, a clear 
increase in the yield was observed with the immobilized porphyrins (Table 6.6). As in 
the PhIO reactions, the immobilized catalyst appeared to be relatively stable toward 
the reaction conditions (Figure 6.7). 



Table 6.6: Conversion of cyclooctene to cyclooctene oxide with 400 |amol cyclooctene 
and 80 )amol H2O2 in ImL of solution using imidazole and porphyrin. 



SA ImODPA/SA MnP4 Cyclooctene oxide 

Yield 



Blank 0.2%-0.3% 

Homogeneous* 0.5%-1.0% 



Films " 12.6%, 12.8% 



Homogeneous solution contains 40 |j,mol ImH and Inmol of MnPO 



The epoxidation of cyclooctene was also run in the presence of excess oxidant 
with approximately the same molar ratio of catalyst (10"^ vs other reactants). The ratio 
examined using excess oxidant was 80 i^mol cyclooctene to 400 jamol H2O2. The 
results were surprising in one aspect. After the 24 hr catalysis reaction was concluded, 
the UV-vis of the film showed nearly complete demetallation of the porphyrin (Figure 
6.8). 



161 



. V -!■'■ 



0.04 



0.03- 



After SA 

rinsing 

catalysis 




350 400 450 500 

Wavelength (nm) 



550 



Figure 6.8: SA ImODPA/SA MnP4 after rinsing and after 24 hr in catalysis reaction 
with excess HjOj. 



Also, the epoxide yield with the film was now around 0.5% - nearly identical to that 
seen in the homogeneous and blank reactions run in vials. Interestingly, when this 
reaction was run in a vial with 1 |imol of MnPO, mimicking the literature procedure, 
the yield was 74%, closely resembling the yield reported with a MnTFPP CI 
catalyst. 116 

6.2.2.2. Oxidation with 10^ times less catalyst vs. other reactants (.8 fimol 
H^Ot or .4 ^mol cvclooctene vs.! nmol MnPl Because the epoxide yield in the above 
reactions may be lower than usual due to the very small amount of catalyst present 
relative to reactants, an attempt was made to bring these concentrations closer to the 
literature ratios of 10:1 reactant to catalyst. Additionally, this oxidant concentration 
(400 lamol vs. 1 nmol catalyst) clearly causes the degradation of the porphyrin films. 



(D 
O 

C 
(0 

SI 
o 

< 



-0.005 



350 



400 



450 



500 



550 



Wavelength (nm) 



162 




600 



Figure 6.9: SA ImODPA/SA MnP4 film with catalysis using 8 i^mol cyclooctene to 
0.2 |j,mol HjOj. 



The solutions were, therefore, diluted 100-fold and similar reactions were 
investigated. The new reaction molar ratio was 8 ^irnol cyclooctene to 0.2 ^mol H2O2 
to 1 nmol of catalyst. The concentration of catalyst in the homogeneous case was set 
by the concentration in the films, which could not be increased. Unfortunately, even 
with excess substrate to protect the porphyrin structure, more serious catalyst 
bleaching was observed in these films than in those with PhIO (Figure 6.9). In 
addition, the epoxide yield with the immobilized catalyst under these conditions was 
similar to or less than that observed in the homogeneous case. The reasons behind 
these results are not clear. 



163 



6.3 Conclusions 

From the above results, it appears that the immobilization of the porphyrins does 
slightly improve the catalytic efficiency of the Mn-porphyrin with PhlO as the 
oxidant. An improvement is also observed in the catalyst stability in the zirconium 
phosphonate films relative to the homogeneous and previous LB film examples. The 
stability and easy recovery of these immobilized porphyrins is an advantage over 
previous literature cases. 

The improvement of the epoxide yields using H2O2 using the immobilized 
porphyrin and imidazole did show promise when the molar ratio was ca. 400 |j,mol 
cyclooctene to 80 j^mol HjOj to 1 nmol of catalyst. Additionally, these films appeared 
to be significantly more stable over 24 hr than the homogeneous catalysts. 
Unfortunately, with excess oxidant, the Mn-porphyrin in the film completely 
demetallates. 

Overall, zirconium phosphonate templates allowed the successful incorporation 
of both catalytic porphyrins and imidazoles into thin film environments. The 
inorganic template also introduced significant film stability and overall improved the 
catalytic efficiency of the immobilized catalysts. 



REFERENCES 

(1) Papkovsky, D. B. Sensors and Acts. B: Chem. 1995, 1-3, 213-218. 

(2) Jung, G. Y., Pearson, C, Kilitziraki, M., Horsburgh, L.E., Monkman, A.P., 
Samuel, I.D.W., Petty, M.C. J. Mater. Chem. 2000, 10, 163-167. 

(3) Ogata, N. Makromol. Chem.-Macromol. Symp. 1992, 53, 191-200. 

(4) Papkovsky, D. B., Ovchinikov, A.N., Ponomarev, G.V., Korpela, T. Anal. Lett. 
1997, 30, 699-719. 

(5) Fabianowski, W., Jaccodine, R., Kodnani, R., Pearson, R., Smektala, P. Adv. 
Mat Opt. Elec. 1995, 5, 199-213. 

(6) Papkovsky, D. B., Desyaterik, I.V., Ponomarev, G.V., Kurochkin, I.N., 
Korpela, T. Anal.Chim.Acta 1995, 310, 233-239. 

(7) Aramaki, K. Corrosion Sci. 1999, 41, 1715-1730. 

(8) Arnold, D. P., Manno, D., Micocci, G., Serra, A., Tepore, A., Valli, L. Thin 
Solid Films 1998, 327-329, 341-344. 

(9) Penza, M., Milella, E., Anisimkin, V.I. IEEE T. Ultrason. Ferr. 1998, 45, 
1125-1132. 

(10) Kauffinan, F., Hoffman, B., Erbach, R., Heiliger, L., Siegmund, H.U., Volker, 
M. Sensors and Acts B-chemical 1994, 18, 60-64. 

(1 1) Marlow, A. L., Davis, J.T. Tet Lett 1999, 40, 3539-3542. 

(12) Petrucci, M. G. L.; Kakkar, A. K. Chem. Mater. 1999, 11, 269-276. 

(13) Suh, J. H., Hong, S.H. J. Am. Chem. Soc. 1998, 120, 12545-12552. 

(14) Agarwal, V. K. Physics Today 1988, 40. 

(15) Roberts, G. G. Langmuir-Blodgett Films; Plenum Press: New York, 1990. 

(16) Langmuir, I. J. Am. Chem. Soc. 1917, 39, 1848-1906. 

(17) Blodgett, K. A.; Langmuir, I. Phys. Rev. 1937, 51, 964. 

(18) Levine, O., Zisman, W.A. J. Phys. Chem. 1957, 61, 1068. 



164 



165 



(19) Bartell, L. S., Ruch, R.J. J. Phys. Chem 1956, 60. 

(20) Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir- 
Blodgett to Self-Assembly; Academic Press: Boston, 1991. 

(21) Zasadzinski, J. A.; Viswanathan, R.; Madsen, L.; Gamaes, J.; Schwartz, D. K. 
Science 1994, 263, 1726-1733. 

(22) Gaines, G. J. Insoluble Monolayers at Liquid-Gas Interfaces; Wiley- 
Interscience: New York, 1966. 

(23) Tredgold, T. H. J. Mater. Chem. 1995, 5, 1095-1 106. 

(24) Aston, M. S. Chem. Soc. Rev. 1993, 67-71. 

(25) Orrit, M.; Mobius, D.; Lehmann, U.; Meyer, H. J. Chem. Phys. 1986, 85, 4966- 
4979. 

(26) Mobius, D.; Orrit, M.; H., G.; Meyer, H. Thin Solid Films 1985, 132, 41-53. 

(27) Drago, R. S. Physical Methods for Chemists; Saunders College Publishing: 
Orlando, 1992. 

(28) Byrd, H.; Pike, J. K.; Talham, D. R. Chem. Mater. 1993, 5, 709-715. 

(29) Byrd, H.; Whipps, S.; Pike, J. K.; Talham, D. R. Thin Solid Films 1994, 244, 
768-771. 

(30) Offord, D. A.; Sachs, S. B.; Ennis, M. S.; A.Eberspacher, T.; Griffin, J. H.; 
Chidsey, C. E. D.; Collman, J. P. J. Am. Chem. Soc. 1998, 120, 4478-4487. 

(31) Practical Surface Analysis; 2nd ed.; Briggs, D.; Seah, M. P., Eds.; John Wiley 
and Sons: Chichester, 1990; Vol. 2. 

(32) Seah, M. P.; Dench, W. A. Surface Interface Analysis 1979, 7,1-11. 

(33) Zhang, Z.; Nakashima, K.; Lai Verma, A.; Yoneyama, M.; Iriyama, K.; Ozaki, 
Y. Langmuir 1998, 14, 1 177-1 182. 

(34) Sagiv, J. J. Am. Chem. Soc. 1980, 102, 92. 

(35) Nuzzo, R. G., Allara, D.L. J. Am. Chem. Soc. 1983, 105, 4481. 

(36) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 
1987, 109, 3559-3568. 

(37) Strong, L., Whitesides, G.M. Langmuir 1988, ^,316. 

(38) Byrd, H.; Whipps, S.; Pike, J. K.; Ma, J.; Nagler, S. E.; Talham, D. R. J. Am. 
Chem. Soc. 1994, 116, 295-301. 



166 



(39) Byrd, H.; Pike, J. K.; Talham, D. R. Syn. Met. 1995, 71, 1977-1980. 

(40) Seip, C. T.; Byrd, H.; Pike, J. K.; Whipps, S.; Talham, D. R. In Physical 
Supramolecular Chemistry; Echegoyen, L., Kaifer, A. E., Eds., 1996; Vol. 485. 

(41) Seip, C. T.; Byrd, H.; Talham, D. R. Inorg. Chem. 1996, 35, 3479-3483. 

(42) Seip, C. T.; Granroth, G. E.; Meisel, M. W.; Talham, D. R. J. Am. Chem. Soc. 
1997, J 19, 7084-7094. 

(43) Clearfield, A. Chem. Mater. 1998, 10, 2801-2810. 

(44) Cemto, G., Trifiro, F., Ebner, J.R., Franchetti, V.M. Chem. Rev. 1988, 88, 55. 

(45) Frink, K. J.; Wang, R.-C; Colon, J. L.; Clearfield, A. Inorganic Chemistry 
1991,50,1438-1441. 

(46) Cao, G.; Mallouk, T. E. Inorganic Chemistry 1991, 30, 1434-1438. 

(47) Cao, G.; Lynch, V. M.; Yacullo, L. N. Chem. Mater. 1993, 5, 1000-1006. 

(48) Drumel, S., Janvier, P., Bujoli-Doeuff, M., Bujoli, B. J. Mater. Chem. 1996, 6, 
1843-1847. 

(49) Kaschak, D. M., Johnson, S.A., Hooks, D.E., Kim, H.-N., Ward, M.D., 
Mallouk, T.E. J. Am. Chem. Soc. 1998, 120, 10887-10894. 

(50) Kraus, K. A., Phillips, H.O. J. Am. Chem. Soc. 1956, 78, 644. 

(51) Alberti, G. Ace. Chem. Res. 1978, 11, 163. 

(52) Clearfield, A. Chem. Rev. 1988, 88, 125-148. 

(53) Clearfield, A. In Progress in Inorganic Chemistry; John Wiley & Sons: NY, 
1998; Vol. 47. 

(54) Cao, G.; Lee, H.; Lynch, V. M.; Mallouk, T. E. Solid State Ionics 1988, 26, 63- 
69. 

(55) Cao, G.; Lee, H.; Lynch, V. M.; Mallouk, T. E. Inorg. Chem. 1988, 27, 2781- 
2785. 

(56) Cunningham, D.; Hennelly, P. J. D. Inorg Chim. Acta 1979, 37, 95-102. 

(57) Zhang, Y.; Scott, K. J.; Clearfield, A. Chemistry of Materials 1993, 5, 495- 
499. 

(58) Cao, G.; Lynch, V. M.; Swinnea, J. S.; Mallouk, T. E. Inorganic Chemistry 
1990,29,2112-2117. 



'■«" 'h "FT'' 



167 



(59) Wang, R.-C; Zhang, Y.; Hu, H.; Frausto, R. R.; Clearfield, A. Chem. Mater. 
1992, 4, 864-870. 

(60) Thomas, L. C; Chittenden, R. A. Spectrochimica Acta 1970, 26A, 781-800. 

(61) Lee, H.; Kepley, L. J.; Hong, H.-G.; Mallouk, T. E. J. Am. Chem. Soc. 1988, 
770,618-620. 

(62) Lee, H.; Kepley, L. J.; Hong, H.-G.; Akhter, S.; Mallouk, T. E. J. Phys. Chem. 
1988, P2, 2597-2601. 

(63) Cao, G.; Hong, H.-G.; Mallouk, T. E. Ace. Chem. Res. 1992, 25, 420-427. 

(64) Fanucci, G. E.; Seip, C. T.; Petruska, M. A.; Ravaine, S.; Nixon, C. M.; 
Talham, D. R. Thin Solid Films 1998, 327-329, 331-335. 

(65) Petruska, M. A.; Fanucci, G. E.; Talham, D. R. Chem. Mater. 1998, 10, \11- 
189. 

(66) Fanucci, G. E.; Petruska, M. A.; Meisel, M. W.; Talham, D. R. J. Solid State 
Chem. 1999,7^5,443-451. 

(67) Petruska, M. A.; Talham, D. R. Chem. Mater 1998, 10, 3673-3682. 

(68) Croney, J. C, Helms, M.K., Jameson, D.M., Larsen, R.W. J. Phys. Chem. B 
2000, 104, 973-977. 

(69) Hsiao, J.-S., Krueger, B.P., Wagner, R.W., Johnson, T.E., Delaney, J.K., 
Mauzerall, D.C., Fleming, G.R., Lindsey, J.S., Bocian, D.F., Donohoe, R.J. J. 
Am. Chem. Soc. 1996,118, 11181-11193. 

(70) Ishida, A.; Sakata, Y.; Majima, T. Chem. Lett. 1998, 267-268. 

(71) Jolliffe, K.; Bell, T.; Ghiggino, K.; Langford, S.; Paddon-Row, M. Angew. 
Chem. Int E. 1998, 37, 916-919. 

(72) Florsheimer, M., Mohwald, H. Thin Solid Films 1988, 759, 1 1 5- 1 23. 

(73) Amini, M. K., Shahrokhian, S., Tangestaninejad, S. Analyst 1999, 124, 1319- 
1322. 

(74) Arnold, D.; Manno, D.; Micocci, A.; Tepore, A.; Valli, L. Langmuir 1997, 13, 
5951-5956. 

(75) Smith, V. C, Batty, S.V., Richardson, T., Foster, K.A., Johnstone, R.A.W., 
Sobral, A.J.F.N., Rocha Gonzales, A.M.dA. Thin Solid Films 1996, 284-285, 
911-914. 

(76) Schenning, A.; Hubert, D.; Feiters, M.; Nolte, R. Langmuir 1996, 12, 1572- 
1577. 



168 



(77) Lahiri, J., Fate, G.D., Ungashe, S.B., Groves, J.T. J. Am. Chem. Soc. 1996, 
/75, 2347-2358. 

(78) Schenning, A. P. H. J., Lutje Spelberg, J.H., Driessen, M.C.PF., Hauser, 
M.J.B., Feiters, M.C., Nolte, R.J.M. J. Am. Chem. Soc. 1995, 117, 12655- 
12656. 

(79) The Porphyrins; Dolphin D. Academic Press: New York, 1978; Vol. lA. 

(80) The Porphyrins; Gouterman, M., Dolphin D. Ed.; Academic Press: New York, 
San Francisco, London, 1978; Vol. III. 

(81) Gouterman, M. J. Mol. Spec. 1961, 6, 138-163. 

(82) Kroon, J. M., Koehorst, R.B.M., van Dijk, M., Sanders, G.M., Sudholter, 
E.J.R. J. Mater. Chem. 1997, 7, 615-624. 

(83) The Porphyrins; Academic Press: New York, 1978; Vol V. 

(84) Kroon, J.; Sudholter, E.; Schenning, A.; Nolte, R. Langmuir 1995, 11, 214- 
220. 

(85) Zhang, J.; Wang, D.; Chen, Y.; Li, T.; Mao, H.; Tian, H.; Zhou, Q.; Xu, H. 
Thin Solid Films 1997, 300, 208-212. 

(86) Gregory, B. W., Vaknin, D., Gray, J.D., Ocko, B.M., Stroeve, P., Cotton, T.M., 
Struve, W.S. J. Phys. Chem. B 1997, 101, 2006-2019. 

(87) Tour, J. M., Jones, LeRoy 111, Pearson, D.L., Lamba, J.J.S., Burgin, T.P., 
Whitesides, G.M., Allara, D.L., Parikh, A.N., Atre, S.V. J. Am. Chem. Soc. 
1995,777,9529-9534. 

(88) Abatti, D.; Zaniquelli, M. E.; lamamoto, Y.; Idemori, Y. Thin Solid Films 
1997, 310, 296-302. 

(89) Anikin, M.; Tkachenko, N.; Lemmetyinen, H. Langmuir 1997, 13, 3002-3008. 

(90) Honeyboume, C; Barrell, K. J. Mater. Chem. 1996, 6, 323-329. 

(91) Ruaudel-Teixier, A.; Barraud, A.; Belbeoch, B.; RouUiay, M. Thin Solid Films 
1983, 99, 33-40. 

(92) Song, X.; Miura, M.; Xu, X.; Taylor, K.; Majumder, S.; Hobbs, J.; Cesarano, 
J.; Shelnutt, J. Langmuir 1996, 12, 2019-2027. 

(93) Breslow, R., Zhang, X., Huang, Y. J. Am. Chem. Soc. 1997, 77P, 4535-4536. 

(94) Robert, A., Loock, B., Momenteau, M., Meunier, B. Inorg. Chem. 1991, 30, 
706-711. 

(95) Beck, M. J., Gopinath, E., Bruice, T.C. J. Am. Chem. Soc. 1993, 775, 21-29. 



169 



(96) Mohajer, D., Monfared, H.H. J. Chem. Res. S 1998, 772-773. 

(97) Arasasingham, R. D., He, G.-X., Bruice, T.C. J. Am. Chem. Soc. 1993, 115, 
7985-7991. 

(98) Arasasingham, R. D. J., S., Bruice, T.C. J. Am. Chem. Soc. 1992, 114, 2536- 

2544. ■ V. :. <., 

(99) Baciocchi, E., Boschi, T., Galli, C, Lapi, A., Tagliatesta, P. Tetrahedron 1997, 
53, 4497-4502. 



(100 
(101 

(102 

(103 

(104 
(105 

(106 

(107 
(108 

(109 
(110 
(111 
(112 
(113 
(114 
(115 



Quo, C.-C, Li, H.-P., Xu, J.-B. J. Cat. 1999, 185, 345-351. 

Collman, J. P., Zhang, X., Lee, V.J., Uffelman, E.S., Brauman, J.I. Science 
1993, 2<57, 1404-1411. 

Lai, T.-S., Kwong, H.-L., Che, C.-M., Peng, S.-M. Chem. Commun. 1997, 

2373-2374. 

Gilmarten, C; Lindsay Smith, J. J. Chem. Soc. Perkin Trans. 1995, 2, 243- 
251. 

Miki, K.; Sato, Y. Bull. Chem. Soc. Jpn. 1993, 66, 2385-2390. 

Deniaud, D.; SchoUom, B.; Mansuy, D.; Rouxel, J.; Battioni, P.; Bujoli, B. 
Chem. Mater. 1995, 7, 995-1000. 

Martinez-Lorente, M. A., Battioni, P., Kleemiss, W., Bartoli, J.F., Mansuy, D. 
J. Mol. Cat. A 1996, 113, 343-353. 

Cooke, P. R., Lindsay Smith, J.R. Tet. Lett. 1992, 33, 2737-2740. 

Neys, P. E. F., Severeyns, A., Vankelecom, I.F.J., Ceulemans, E., Dehaen, W., 
Jacobs, P.A. J. Mol. Cat. A Chemical 1999, 144, 1>11>-?>11. 

Groves, J. T., Ungashe, S.B. J. Am. Chem. Soc. 1990, 112, 7796-7797. 

Campestrini, S., Meunier, B. Inorg. Chem. 1992, 31, 1999-2006. 

Yuan, L.-C, Bruice, T.C. J. Am. Chem. Soc. 1986, 108, 1643-1650. 

Tangestaninejad, S., Moghadam, M. Syn. Comm. 1998, 28, 427-432. 

Coyle, C. L., Rafson, P.A., Abbott, E.H. Inorg. Chem. 1973, 12, 2007-2010. 

Arasasingham, R. D., Bruice, T.C. J. Am. Chem. Soc. 1991, 113, 6095-6103. 

Boucher Coord. Chem. Rev. 1972, 7, 289-329. 



170 



:i 16) Battioni, P., Renaud, J.P., Bartoli, J.F., Reina-Artiles, M., Fort, M., Mansuy, D. 
J. Am. Chem.Soc 1988, 110, 8462-8470. 

:il7) Kern, W. J. Electrochem. Soc. 1990, 137, 1887-1892. 

fl 18) Advincula, R. Dissertation, University of Florida, 1994. 

fl 19) Evans, D. F. J. Chem. Soc. 1950, 2003. . .. :. 

fl20) Schubert, E. M. J. Chem. Ed. 1992, 69, 62. - 

[\2\) Saltzman, H., Sharefkin, J.G. Organic Synthesis , 43, 60-61. 

:i22) Lucas, H. J., Kennedy, E.R. Organic Synthesis , Coll. Vol. 3, 483, 1955. 

'123) Montanari, F. Metalloporphyrins Catalysed Oxidations; Kluwer Academic 
Publishers, 1994. 

:i24) Caughey, W.; Deal, R.; Weiss, C; Goutermann, M. J. Mol. Spec. 1965, 16, 
451-463. 

:i25) Talham, D. R.; Seip, C. T.; Whipps, S.; Fanucci, G. E.; Petruska, M. A.; Byrd, 
H. Comments Inorg Chem. 1997, 19, 133-151. 

;i26) Petruska, M. A., Talham, D.R. Langmuir 2000, submitted. 

[\21) Day, V. W., Stults, B.R., Tasset, E.L., Day, R.O., MarianeUi, R.S. J. Am. 
Chem. Soc. 1974, 96, 2650-2652. 

:i28) Tulinsky, A., Chen, B.M.L. J. Am. Chem. Soc. 1977, 99, 3647-3651. 

:i29) Hansen, A. P., Goff, H.M. Inorg Chem. 1984, 23, 4519-4525. 

:i30) Schardt, B. C, Hollander, F.J., Hill, C.L. J. Am. Chem. Soc. 1982, 104, 3964- 
3972. 

:i31) Boucher, L. J. J. Am. Chem. Soc. 1970, 92, 2725-2730. 

^32) Powell, M. F., Pai, E.F., Bruice, T.C. J. Am. Chem. Soc. 1984, 106, 3277-3285. 

:i33) Mu, X. H., Schultz, F.A. Inorg Chem. 1992, 31, 3351-3357. 

:i34) Maiti, N. C, Mazumdar, S., Periasamy, N. J. Phys. Chem. B 1998, 102, 1528- 
1538. 

:i35) Nixon, C. M.; Claire, K. L.; Odobel, F.; Bujoli, B.; Talham, D. R. Chem. 
Mater. 1999, 11, 965-976. 



BIOGRAPHICAL SKETCH 

Christine M. Nixon Lee was bora in Dayton, Ohio, on Febraary 17, 1973. She 
moved with her family to Mansfield, Ohio, in fifth grade, and she graduated fi-om 
Lexington High School in June of 1991. Christine entered Baldwin- Wallace College 
in Berea, Ohio, in September 1991 as a music major, and in the fall of 1992, she began 
her chemistry major. Christine graduated from Baldwin- Wallace College summa cum 
laude in June of 1995 with a B.S. in chemistry and a minor in music. 

In the summer of 1995, she started studying at the University of Florida, and 
joined Dr. Dan Talham's group in the spring of 1997-just in time to do her oral 
qualifying exam. She married Lawrence Lee in July of 1999, and will be joining him 
m New York City after she graduates in the spring of 2000 with her Ph.D. in 
chemistry. 



171 



I certify that I have read this study and that in my opinion it conforms to acceptable 
standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation 
for the degree of Doctor of Philosophy. 




Daniel R. Talham, Chairman 
Associate Professor of Chemistry 



I certify that I have read this study and that in my opinion it conforms to acceptable 
standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation 
for the degree of Doctor of Philosophy. 



William Weltner ' 
Professor of Chemistry 



I certify that I have read this study and that in my opinion it conforms to acceptable 
standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation 
for the degree of Doctor of Philosophy. 




^A-^ 



Martin Vala 
Professor of Chemistry 

I certify that I have read this study and that in my opinion it conforms to acceptable 
standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation 
for the degree of Doctor of Philosophy. 




fu. (^>-< 



A^IU, i/^~^^<^pUA- 



Kenneth Wagener 
Professor of Chemistry 

I certify that I have read this study and that in my opinion it conforms to acceptable 
standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation 
for the degree of Doctor of Philosophy. 



David P. Chynoweth \ 

Professor of Agricultural and Biological 

Engineering 



This dissertation was submitted to the Graduate Faculty of the Department of 
Chemistry in the College of Liberal Arts and Sciences and to the Graduate School and was 
accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. 

August 2000 ^ 



Dean, Graduate School 



LD 

1780 

20_oo 






< . -. 7 r. 



UNIVERSITY OF FLORIDA 



3 1262 08555 1777